INDEX

Explanations

occurrences of the character 'e'

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

st

-0.61

stuffs

-0.58

nt

-0.57

Tikang

-0.55

 سكانية

-0.54

IVEREF

-0.54

matchCondition

-0.54

ApiModelProperty

-0.53

 unknownFields

-0.52

 lagar

-0.52

POSITIVE LOGITS

0.64

0.60

tvguidetime

0.60

printStackTrace

0.58

ьаж

0.56

oporosis

0.54

orszá

0.54

chedelic

0.54

jectable

0.52

entary

0.50

Activations Density 0.153%

occurrences of the character 'e'

No Comments

No Known Activations