INDEX

Explanations

references to the number "90"

oai_token-act-pair · gpt-3.5-turbo

references to the number 90

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

lay

-0.81

seeing

-0.79

board

-0.78

wan

-0.76

lish

-0.75

untu

-0.73

pocket

-0.72

think

-0.70

lings

-0.69

mine

-0.69

POSITIVE LOGITS

%:

1.03

%"

1.00

0.99

 percent

0.98

Âº

0.94

%;

0.93

 degrees

0.92

mph

0.92

Â°

0.91

%,

0.88

Activations Density 0.067%

references to the number "90"

references to the number 90

No Comments

No Known Activations