INDEX

Explanations

terms related to burning and fire

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

ViewFeatures

-0.78

 ujednoznacz

-0.65

 تانيه

-0.63

+#+#

-0.62

 חיצוניים

-0.62

stateProvider

-0.61

 StatefulWidget

-0.60

Autoritní

-0.59

okhttp

-0.59

dini

-0.57

POSITIVE LOGITS

 burn

0.89

 burns

0.89

 burned

0.84

 BURN

0.82

burning

0.82

 burning

0.82

 burnt

0.80

 Burn

0.80

 Burns

0.80

burnt

0.77

Activations Density 0.074%

terms related to burning and fire

No Comments

No Known Activations