INDEX

Explanations

specific examples or mentions related to technology or names, possibly programming languages

oai_token-act-pair · gpt-3.5-turbo

sequences of letters or character patterns that resemble technical or code-like terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 predec

-0.82

imeters

-0.68

 calories

-0.64

 Thumbnails

-0.64

achev

-0.64

agascar

-0.61

uity

-0.60

 cryst

-0.59

 artif

-0.58

avorite

-0.58

POSITIVE LOGITS

 Rowling

0.91

EStream

0.90

 Marriott

0.80

eki

0.75

dan

0.73

EStreamFrame

0.73

owicz

0.70

ansen

0.69

 Sabha

0.69

ohan

0.68

Activations Density 0.281%

specific examples or mentions related to technology or names, possibly programming languages

sequences of letters or character patterns that resemble technical or code-like terms

No Comments

No Known Activations