INDEX

Explanations

locations in California and Florida

oai_token-act-pair · gpt-3.5-turbo

references to locations, primarily states in the U.S

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Crimean

-0.78

BOOK

-0.73

buster

-0.70

 foot

-0.66

road

-0.62

 pointers

-0.61

Proof

-0.60

 ration

-0.60

 Gravity

-0.60

isation

-0.59

POSITIVE LOGITS

.,

1.24

jer

1.02

.;

0.98

olla

0.90

achu

0.90

adelphia

0.88

atche

0.82

.?

0.81

aii

0.81

./

0.80

Activations Density 0.027%

locations in California and Florida

references to locations, primarily states in the U.S

No Comments

No Known Activations