INDEX

Explanations

references to achieving success or successful outcomes

oai_token-act-pair · gpt-3.5-turbo

phrases related to achieving success or failure 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

inki

-0.87

 beds

-0.71

isted

-0.67

ï¸ı

-0.66

Font

-0.66

pic

-0.65

wi

-0.62

Zone

-0.61

ferred

-0.60

 opin

-0.60

POSITIVE LOGITS

 miser

0.88

successfully

0.81

 rollout

0.75

 impe

0.75

tein

0.75

 teamwork

0.73

 successfully

0.73

 midterm

0.72

 muster

0.71

 balancing

0.71

Activations Density 0.315%

references to achieving success or successful outcomes

phrases related to achieving success or failure in various contexts

No Comments

No Known Activations