INDEX

Explanations

codes or acronyms related to specific organizations or places

oai_token-act-pair · gpt-3.5-turbo

references to specific organizations or entities indicated by "DF."

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

sole

-0.78

 Metatron

-0.70

stakes

-0.66

bell

-0.66

 amen

-0.64

 bells

-0.63

Answer

-0.62

comes

-0.60

swick

-0.60

frames

-0.60

POSITIVE LOGITS

DF

0.95

ried

0.94

erm

0.93

rost

0.92

avorite

0.92

WD

0.90

amily

0.89

actory

0.89

sg

0.87

iants

0.86

Activations Density 0.016%

codes or acronyms related to specific organizations or places

references to specific organizations or entities indicated by "DF."

No Comments

No Known Activations