INDEX

Explanations

numerical values within a text, particularly quantities or statistics

oai_token-act-pair · gpt-3.5-turbo

numerical representations or statistics related to varying contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 swearing

-0.69

£ı

-0.66

 vanity

-0.65

 descending

-0.64

 alienation

-0.61

 Militia

-0.61

 Lash

-0.60

 convey

-0.60

assetsadobe

-0.59

 vested

-0.59

POSITIVE LOGITS

Â½

1.23

0.98

PO

0.94

 Â½

0.93

0.91

0.90

nd

0.90

0.88

Activations Density 0.191%

numerical values within a text, particularly quantities or statistics

numerical representations or statistics related to varying contexts

No Comments

No Known Activations