INDEX

Explanations

technical terms related to the functionality and operation of systems or devices

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "work" 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 @ 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

ylon

-0.82

anamo

-0.79

 Flavoring

-0.76

ilings

-0.71

antha

-0.70

 Shal

-0.69

idium

-0.68

ModLoader

-0.62

ensor

-0.60

fty

-0.60

POSITIVE LOGITS

flows

1.01

bench

0.96

station

0.88

 ethic

0.87

heet

0.87

manship

0.86

river

0.79

hops

0.79

horse

0.76

 overtime

0.76

Activations Density 0.074%

technical terms related to the functionality and operation of systems or devices

occurrences of the term "work" in various contexts

No Comments

No Known Activations