INDEX

Explanations

terms related to classified or sensitive information or documents

oai_token-act-pair · gpt-3.5-turbo

references to classified and sensitive information

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

STRUCT

-0.76

ocal

-0.70

stad

-0.70

Sund

-0.68

Stud

-0.63

Vel

-0.62

Econom

-0.60

ction

-0.59

 Sadd

-0.59

STER

-0.58

POSITIVE LOGITS

 relating

0.91

 leaked

0.89

 stored

0.85

 glean

0.84

 trove

0.83

 intercepted

0.81

 pertaining

0.81

 obtained

0.79

 incrim

0.78

 unlawfully

0.78

Activations Density 0.161%

terms related to classified or sensitive information or documents

references to classified and sensitive information

No Comments

No Known Activations