INDEX

Explanations

language suggesting the potential to improve operations or processes

oai_token-act-pair · gpt-3.5-turbo

phrases indicating capability or potential

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Fighter

-0.68

edient

-0.65

 rehearsal

-0.64

 striving

-0.62

 Generation

-0.62

 honoring

-0.61

 revision

-0.60

 guarding

-0.60

Moz

-0.59

 Gadget

-0.59

POSITIVE LOGITS

't

1.48

adian

1.23

berra

1.18

NOT

1.03

vas

0.98

 attest

0.94

tera

0.92

isters

0.90

thus

0.87

 afford

0.84

Activations Density 0.176%

language suggesting the potential to improve operations or processes

phrases indicating capability or potential

No Comments

No Known Activations