INDEX

Explanations

statements regarding established facts or assertions

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

the fact that

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

mergeFrom

-0.54

nexpected

-0.53

tershire

-0.51

///

-0.47

nima

-0.46

OSP

-0.46

terase

-0.46

ırken

-0.45

bibfnamefont

-0.45

 Sommers

-0.45

POSITIVE LOGITS

 fact

1.15

fact

0.96

Fact

0.95

 Fact

0.94

 Tatsache

0.85

 FACT

0.85

 faptul

0.81

 Facts

0.75

 facts

0.69

FACT

0.66

Activations Density 0.018%

statements regarding established facts or assertions

the fact that

No Comments

No Known Activations