INDEX

Explanations

statements regarding questioning and responses in a conversational context

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

Follows "when" and precedes a question

np_acts-logits-general · gemini-2.0-flash

when asked

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

didSet

-0.81

UserScript

-0.80

 فريبيس

-0.72

featureID

-0.71

 itſelf

-0.70

 ddelweddau

-0.69

 дописавши

-0.69

parsedMessage

-0.69

photobucket

-0.69

 PeEnEo

-0.68

POSITIVE LOGITS

 asked

1.95

Asked

1.72

 Asked

1.67

asked

1.35

 questioned

1.26

 inquired

1.16

 queried

1.09

 question

1.08

Question

1.07

 pregunt

1.06

Activations Density 0.184%

statements regarding questioning and responses in a conversational context

Follows "when" and precedes a question

when asked

No Comments

No Known Activations