INDEX

Explanations

dates formatted like "1st November"

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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 histories

-0.69

 boundaries

-0.64

 devils

-0.63

 angles

-0.61

folk

-0.60

 mascul

-0.60

 hygiene

-0.60

izations

-0.59

olds

-0.59

gotten

-0.59

POSITIVE LOGITS

st

1.36

Password

1.27

1.25

1.20

Â½

1.02

0.99

âģ

0.95

0.93

RM

0.87

Activations Density 0.098%

dates formatted like "1st November"

occurrences of the number '1'

No Comments

No Known Activations