INDEX

Explanations

locations described with street names

oai_token-act-pair · gpt-3.5-turbo

references to specific locations or streets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

arious

-0.83

fulness

-0.81

ERAL

-0.81

berman

-0.80

itious

-0.77

aldi

-0.76

lance

-0.75

figure

-0.74

essee

-0.72

league

-0.71

POSITIVE LOGITS

NW

0.92

 Avenue

0.86

NE

0.85

 corridor

0.82

 Armory

0.81

 Corridor

0.81

 subway

0.79

 East

0.78

 intersection

0.77

 Bridge

0.76

Activations Density 0.027%

locations described with street names

references to specific locations or streets

No Comments

No Known Activations