INDEX

Explanations

phrases related to written documents or papers

oai_token-act-pair · gpt-3.5-turbo

references to academic papers or publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

cffffcc

-0.77

Ģ

-0.73

obal

-0.73

 ambassadors

-0.70

ostic

-0.69

ostics

-0.68

ively

-0.67

oise

-0.67

ivals

-0.66

ois

-0.66

POSITIVE LOGITS

clip

1.30

 towels

1.14

clips

1.09

 towel

0.98

backs

0.97

pus

0.96

bag

0.95

pee

0.90

weight

0.88

craft

0.87

Activations Density 0.037%

phrases related to written documents or papers

references to academic papers or publications

No Comments

No Known Activations