INDEX

Explanations

phrases indicating regret or reflection

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

sentence endings and new clauses

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mybatisplus

-0.38

 saling

-0.29

GrantedAuthority

-0.28

tra

-0.28

endmodule

-0.26

 umož

-0.26

 kiin

-0.26

 toallas

-0.26

CharField

-0.25

ケート

-0.25

POSITIVE LOGITS

 MainAxisSize

0.79

 ब्रेकडाउन

0.69

帖最后由

0.65

 himſelf

0.63

 ModelExpression

0.59

 surla

0.57

__':

0.56

himself

0.55

 AssemblyTitle

0.55

Cyfarwyddwr

0.55

Activations Density 0.069%

phrases indicating regret or reflection

sentence endings and new clauses

No Comments

No Known Activations