INDEX

Explanations

phrases that indicate purpose or intent, often starting with "in order to" or "for."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

tvguidetime

-0.84

PerformLayout

-0.82

 ſtate

-0.79

 pleaſure

-0.79

 myſelf

-0.78

 CreateTagHelper

-0.78

 FetchType

-0.78

 houſe

-0.77

couvrez

-0.75

 ftate

-0.73

POSITIVE LOGITS

0.61

’

0.54

0.52

JoinColumn

0.50

faq

0.50

ندما

0.49

 cual

0.48

するために

0.48

っている

0.47

 côtes

0.47

Activations Density 0.113%

phrases that indicate purpose or intent, often starting with "in order to" or "for."

No Comments

No Known Activations