INDEX

Explanations

mentions of locations, particularly New York (NY)

oai_token-act-pair · gpt-3.5-turbo

references to New York

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

backer

-0.84

tackle

-0.76

module

-0.75

correct

-0.72

balls

-0.72

quel

-0.71

icio

-0.70

tenance

-0.70

cies

-0.69

fitting

-0.68

POSITIVE LOGITS

RB

1.07

SE

0.95

NY

0.89

atana

0.84

 Mellon

0.81

XT

0.80

PD

0.79

RR

0.77

oho

0.75

GW

0.75

Activations Density 0.008%

mentions of locations, particularly New York (NY)

references to New York

No Comments

No Known Activations