INDEX

Explanations

intense expressions of dislike or hatred.

oai_token-act-pair · gpt-4o-mini Triggered by @bot

phrases related to strong negative emotions.

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

expressions of strong dislike or aversion, particularly focusing on the word "hate" and similar negative sentiments.

oai_token-act-pair · claude-3-5-sonnet-20240620 Triggered by @bot

Expressing dislike or aversion

np_acts-logits-general · gemini-2.0-flash

like or hate

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DockStyle

-0.82

WriteAttribute

-0.75

帖最后由

-0.71

GIVEREF

-0.66

клопе

-0.63

Personendaten

-0.61

{@

-0.60

 متعلقه

-0.59

новништво

-0.59

)_/¯

-0.59

POSITIVE LOGITS

 being

0.84

being

0.72

Being

0.71

 Being

0.63

 BEING

0.60

 losing

0.59

 living

0.54

被人

0.54

incon

0.54

 admitting

0.53

Activations Density 0.192%

intense expressions of dislike or hatred.

phrases related to strong negative emotions.

expressions of strong dislike or aversion, particularly focusing on the word "hate" and similar negative sentiments.

Expressing dislike or aversion

like or hate

No Comments

No Known Activations