INDEX

Explanations

instances of numerical figures typically found in references or lists

oai_token-act-pair · gpt-3.5-turbo

mentions of numerical values, particularly those related to measurements or quantities

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

ktop

-0.80

ĸļ

-0.76

prises

-0.76

nesday

-0.75

chio

-0.74

hare

-0.71

ortex

-0.71

akening

-0.71

atile

-0.70

orable

-0.69

POSITIVE LOGITS

1.09

 ILCS

0.88

th

0.83

rd

0.81

mm

0.76

Â¯Â¯Â¯Â¯

0.76

 Centauri

0.75

TH

0.74

nm

0.72

strap

0.69

Activations Density 0.021%

instances of numerical figures typically found in references or lists

mentions of numerical values, particularly those related to measurements or quantities

No Comments

No Known Activations