INDEX

Explanations

numerical patterns in the format "X-X" where X represents a number

oai_token-act-pair · gpt-3.5-turbo

occurrences of the number "1"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Lumpur

-0.87

awaru

-0.79

vel

-0.72

apon

-0.72

 roofs

-0.71

 hygiene

-0.65

usalem

-0.65

 shopping

-0.65

ategory

-0.64

 travelling

-0.64

POSITIVE LOGITS

Password

1.20

1.19

1.14

1.09

1.03

1.02

00000000

0.94

0.93

st

0.91

 Corinthians

0.87

Activations Density 0.054%

numerical patterns in the format "X-X" where X represents a number

occurrences of the number "1"

No Comments

No Known Activations