INDEX

Explanations

references to scientific studies or biomedical terms

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

First word of references

np_acts-logits-general · gemini-2.0-flash

names ending in i, ghi, o, anski

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

[toxicity=0]

-0.63

 alike

-0.51

λίου

-0.51

®.

-0.48

™.

-0.48

-0.45

 فريبيس

-0.45

 therein

-0.45

 therefor

-0.42

GMENT

-0.42

POSITIVE LOGITS

 ***!

0.97

])),

0.92

 nahilalakip

0.89

 للاسماء

0.88

SourceChecksum

0.86

 ProtoMessage

0.85

rungsseite

0.84

)))),

0.82

MessageOf

0.81

LookAnd

0.80

Activations Density 1.380%

references to scientific studies or biomedical terms

First word of references

names ending in i, ghi, o, anski

No Comments

No Known Activations