INDEX

Explanations

instances of the word "When" indicating temporal contexts or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 39-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_39/width_16k/average_l0_131

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.39.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

them

-0.48

 honom

-0.46

 ujednoznacz

-0.40

him

-0.39

 himſelf

-0.38

ihn

-0.37

Them

-0.36

../

-0.36

 adalah

-0.35

 them

-0.34

POSITIVE LOGITS

 confronted

1.12

 faced

1.09

 compared

0.98

 asked

0.97

 they

0.94

we

0.88

 dealing

0.87

 considering

0.87

 viewed

0.84

 discussing

0.84

Activations Density 0.196%

instances of the word "When" indicating temporal contexts or conditions

No Comments

No Known Activations