INDEX

Explanations

terms related to effectiveness or efficiency

oai_token-act-pair · gpt-3.5-turbo

references to various forms of effectiveness or successful strategies

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

forest

-0.74

yg

-0.70

umble

-0.69

pper

-0.67

Hop

-0.65

 Alive

-0.65

fty

-0.60

 Shoes

-0.59

¦

-0.59

ploma

-0.59

POSITIVE LOGITS

 altru

1.01

iveness

0.86

ative

0.83

atives

0.83

abulary

0.83

rations

0.80

nesses

0.79

acies

0.78

 coping

0.75

 deterrent

0.74

Activations Density 0.039%

terms related to effectiveness or efficiency

references to various forms of effectiveness or successful strategies

No Comments

No Known Activations