INDEX

Explanations

statements or declarations related to various situations or topics

oai_token-act-pair · gpt-3.5-turbo

statements or official communications related to events or incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 theorem

-0.65

omics

-0.65

hiba

-0.65

skill

-0.64

EStream

-0.63

oleon

-0.63

otrop

-0.63

atron

-0.63

opian

-0.62

çļ

-0.59

POSITIVE LOGITS

 statement

1.13

Statement

1.03

 clar

1.02

 clarified

1.01

 spokesperson

0.97

 disav

0.95

 clarification

0.94

 refuted

0.94

 apologised

0.93

 emailed

0.92

Activations Density 0.500%

statements or declarations related to various situations or topics

statements or official communications related to events or incidents

No Comments

No Known Activations