INDEX

Explanations

specific mentions of names, titles, or entities

oai_token-act-pair · gpt-3.5-turbo

occurrences of specific high-frequency or significant names and terms relevant to current events or notable entities

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

IFrame

Link

Negative Logits

bender

-0.67

 autistic

-0.63

unc

-0.59

 Takeru

-0.58

 lest

-0.58

 commissions

-0.58

 scissors

-0.56

 Pixar

-0.56

ogle

-0.55

orc

-0.54

POSITIVE LOGITS

bilt

1.17

emort

0.90

export

0.82

Leaks

0.74

ennes

0.69

sov

0.69

amins

0.67

sky

0.67

oÄŁ

0.67

idis

0.67

Activations Density 0.428%

specific mentions of names, titles, or entities

occurrences of specific high-frequency or significant names and terms relevant to current events or notable entities

No Known Activations