INDEX

Explanations

information related to rankings or positions in a list

oai_token-act-pair · gpt-3.5-turbo

rankings and statistics related to performance in various categories

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

tun

-0.62

rod

-0.61

 interns

-0.59

lot

-0.58

WER

-0.57

matter

-0.57

Done

-0.57

CHAT

-0.56

rett

-0.56

Report

-0.56

POSITIVE LOGITS

ibaba

0.86

offensive

0.78

roads

0.71

 overall

0.70

ordinate

0.68

clusively

0.68

 order

0.68

 Honolulu

0.67

 Hawaii

0.67

clusions

0.65

Activations Density 0.092%

information related to rankings or positions in a list

rankings and statistics related to performance in various categories

No Comments

No Known Activations