INDEX

Explanations

punctuation marks, particularly commas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 betweenstory

-0.56

 faſt

-0.52

{*}

-0.47

 Huguen

-0.46

 fertil

-0.46

 fertiliser

-0.46

RAM

-0.45

 afghan

-0.45

 noDo

-0.44

 militar

-0.44

POSITIVE LOGITS

 derfor

0.65

发表于

0.62

corazón

0.56

 therefore

0.55

 eikä

0.54

 soñ

0.54

 bahwa

0.54

 Derfor

0.53

จึง

0.53

therefore

0.52

Activations Density 0.007%

punctuation marks, particularly commas

No Comments

No Known Activations