INDEX

Explanations

words related to various activities or objects in different scenarios

oai_token-act-pair · gpt-3.5-turbo

specific nouns and terms related to events, projects, and organizational contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

srf

-0.71

ecause

-0.67

Ĭ±

-0.59

Ô

-0.57

ĪĴ

-0.55

ilver

-0.55

ĺħ

-0.55

vae

-0.54

ourke

-0.54

cale

-0.53

POSITIVE LOGITS

iest

0.79

 itself

0.74

ieth

0.66

liest

0.66

's

0.65

osphere

0.60

ultimate

0.58

washer

0.57

atta

0.55

iverse

0.55

Activations Density 0.752%

words related to various activities or objects in different scenarios

specific nouns and terms related to events, projects, and organizational contexts

No Comments

No Known Activations