INDEX

Explanations

examples for comparison or illustration

oai_token-act-pair · gpt-3.5-turbo

references to examples or cases in a discussion

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

ormal

-0.80

ogun

-0.72

ves

-0.71

ements

-0.69

sil

-0.69

inal

-0.69

 Flavoring

-0.69

orously

-0.69

ELY

-0.69

 bang

-0.67

POSITIVE LOGITS

©¶æ¥µ

0.79

 owing

0.76

DonaldTrump

0.72

herty

0.69

 Schn

0.67

"@

0.66

 Katz

0.63

 Skydragon

0.63

 subp

0.63

,...

0.63

Activations Density 0.027%

examples for comparison or illustration

references to examples or cases in a discussion

No Comments

No Known Activations