INDEX

Explanations

terms related to people's names

oai_token-act-pair · gpt-3.5-turbo

names and terms related to specific individuals and groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

FORMATION

-0.79

pring

-0.78

å£«

-0.76

chnology

-0.74

zona

-0.73

 explan

-0.71

VOL

-0.69

Interstitial

-0.68

UTION

-0.68

machine

-0.67

POSITIVE LOGITS

anamo

0.86

ablishment

0.79

acle

0.77

 Palestin

0.74

ging

0.73

acles

0.73

Osw

0.72

ARGET

0.71

anyahu

0.69

ometry

0.69

Activations Density 0.021%

terms related to people's names

names and terms related to specific individuals and groups

No Comments

No Known Activations