INDEX

Explanations

descriptions or evaluations of experiences and events

oai_token-act-pair · gpt-3.5-turbo

descriptive phrases and key subjects related to experience and outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 people

-0.50

sites

-0.48

players

-0.46

 claimants

-0.45

 individuals

-0.45

 People

-0.45

 players

-0.44

Companies

-0.44

 guys

-0.44

 booths

-0.44

POSITIVE LOGITS

ãĤ´ãĥ³

0.81

ĸļ

0.77

raising

0.60

interstitial

0.57

atically

0.57

andering

0.57

 transforming

0.54

lasting

0.54

changing

0.53

bda

0.53

Activations Density 0.813%

descriptions or evaluations of experiences and events

descriptive phrases and key subjects related to experience and outcomes

No Comments

No Known Activations