INDEX

Explanations

phrases related to different approaches or methods

oai_token-act-pair · gpt-3.5-turbo

mentions of different strategies or methods used in various contexts

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

pite

-0.80

cell

-0.76

inders

-0.74

 istg

-0.71

cop

-0.69

 isot

-0.67

rib

-0.66

interrupted

-0.65

circ

-0.65

reported

-0.64

POSITIVE LOGITS

 approach

1.12

 Approach

1.11

 approaches

0.93

yip

0.83

ACY

0.78

isons

0.77

ahime

0.76

aphael

0.75

 Ceres

0.75

ctl

0.75

Activations Density 0.013%

phrases related to different approaches or methods

mentions of different strategies or methods used in various contexts

No Comments

No Known Activations