INDEX

Explanations

references to specific quantities or degrees

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 respective

-0.61

 respectively

-0.54

//});

-0.53

^(@)

-0.51

 respectivement

-0.48

ともに

-0.48

harapkan

-0.48

ſelf

-0.48

共に

-0.48

libft

-0.47

POSITIVE LOGITS

 nice

1.00

MLLoader

0.98

bit

0.97

lot

0.94

 really

0.92

 weird

0.87

few

0.85

 little

0.84

really

0.83

WithIOException

0.83

Activations Density 0.289%

references to specific quantities or degrees

No Comments

No Known Activations