INDEX

Explanations

phrases indicating a statement or belief

oai_token-act-pair · gpt-3.5-turbo

assertions and claims made in a discussion or analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

MpServer

-0.77

ãĤ¼ãĤ¦ãĤ¹

-0.72

SourceFile

-0.67

Scroll

-0.66

ipers

-0.65

apsed

-0.64

sie

-0.63

Tank

-0.63

lled

-0.62

mouth

-0.62

POSITIVE LOGITS

 policymakers

0.90

 proponents

0.85

 perceptions

0.74

 progressives

0.74

 historians

0.73

 economists

0.73

 although

0.71

 interpreting

0.70

 comparisons

0.69

 theorists

0.69

Activations Density 0.473%

phrases indicating a statement or belief

assertions and claims made in a discussion or analysis

No Comments

No Known Activations