INDEX

Explanations

keywords related to improvement or enhancement

oai_token-act-pair · gpt-3.5-turbo

references to improvement or enhancement

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

idon

-0.72

mad

-0.67

NH

-0.66

eur

-0.65

Ber

-0.63

Py

-0.62

DP

-0.61

Ar

-0.60

MIT

-0.60

heter

-0.60

POSITIVE LOGITS

 suited

0.95

 behaved

0.84

 than

0.82

 cannabin

0.74

jee

0.73

 lapt

0.73

 behav

0.72

 manag

0.70

 payoff

0.66

 likelihood

0.66

Activations Density 0.023%

keywords related to improvement or enhancement

references to improvement or enhancement

No Comments

No Known Activations