INDEX

Explanations

names of people

oai_token-act-pair · gpt-3.5-turbo

terms related to evaluation and analysis of information or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 unsu

-0.68

kees

-0.66

 favourites

-0.65

 targ

-0.63

 prized

-0.62

heid

-0.61

clud

-0.61

ande

-0.61

ga

-0.61

 Selection

-0.60

POSITIVE LOGITS

ivan

2.92

matic

2.24

ilar

2.19

matically

1.59

matical

1.14

 mileage

1.01

 Piper

0.97

 sober

0.96

ogical

0.95

 Wilderness

0.95

Activations Density 0.055%

names of people

terms related to evaluation and analysis of information or contexts

No Comments

No Known Activations