INDEX

Explanations

phrases related to release notes or documentation

oai_token-act-pair · gpt-3.5-turbo

references to "release notes."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

LOS

-0.79

ãĥĩ

-0.67

nom

-0.66

 displ

-0.64

 waged

-0.62

slaught

-0.62

ILLE

-0.60

namese

-0.60

 Pratt

-0.60

 outp

-0.60

POSITIVE LOGITS

notes

0.98

 notes

0.95

 Notes

0.92

books

0.88

 note

0.83

ppelin

0.79

ense

0.77

Pad

0.75

book

0.74

 Languages

0.73

Activations Density 0.014%

phrases related to release notes or documentation

references to "release notes."

No Comments

No Known Activations