INDEX

Explanations

locations and features related to places and structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

Efq

-0.97

 Jefus

-0.94

 myſelf

-0.94

 himſelf

-0.93

 whoſe

-0.91

StructEnd

-0.90

 Monfieur

-0.89

 reaſon

-0.89

 Chrift

-0.88

 chofe

-0.88

POSITIVE LOGITS

 where

0.66

where

0.56

 pure

0.55

 Where

0.53

mat

0.53

Where

0.51

 multi

0.49

 lots

0.48

 sign

0.48

 waar

0.47

Activations Density 0.413%

locations and features related to places and structures

No Comments

No Known Activations