INDEX

Explanations

numbers at the start of a line or phrase

oai_token-act-pair · gpt-3.5-turbo

numerical values and references to voting data or rankings

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

aten

-0.71

advertising

-0.70

 Herz

-0.65

RG

-0.61

oros

-0.61

idden

-0.60

è»

-0.59

Tour

-0.59

 strugg

-0.59

 proxy

-0.59

POSITIVE LOGITS

uther

0.66

ucer

0.64

 Andromeda

0.62

pee

0.61

pc

0.60

 Cinnamon

0.60

ooks

0.58

nesia

0.57

 Lawn

0.57

Î¼

0.57

Activations Density 0.265%

numbers at the start of a line or phrase

numerical values and references to voting data or rankings

No Comments

No Known Activations