INDEX

Explanations

words related to various problems or issues

oai_token-act-pair · gpt-3.5-turbo

references to issues and challenges

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

rib

-0.84

rica

-0.78

tein

-0.77

alty

-0.75

uclear

-0.74

cer

-0.74

thouse

-0.74

VO

-0.72

ron

-0.71

ewitness

-0.69

POSITIVE LOGITS

 problems

1.16

 plag

1.09

 Problems

1.01

Problem

0.97

hooting

0.89

 headaches

0.85

 difficulties

0.84

 stemming

0.83

 troubles

0.83

 problem

0.80

Activations Density 0.018%

words related to various problems or issues

references to issues and challenges

No Comments

No Known Activations