INDEX

Explanations

phrases related to mechanisms, processes, and actions

oai_token-act-pair · gpt-3.5-turbo

terms related to methods or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

uden

-0.72

 Parables

-0.71

hiro

-0.69

olk

-0.68

atana

-0.68

 Cardinals

-0.66

ield

-0.66

minster

-0.66

sky

-0.65

ouk

-0.65

POSITIVE LOGITS

 mechanism

1.00

 Mechan

0.94

 mechanisms

0.94

utics

0.92

witz

0.85

 whereby

0.83

ptoms

0.81

 symmetry

0.76

 workings

0.74

ality

0.72

Activations Density 0.018%

phrases related to mechanisms, processes, and actions

terms related to methods or processes

No Comments

No Known Activations