INDEX

Explanations

locations

oai_token-act-pair · gpt-3.5-turbo

geographical locations and their associated coordinates or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 brackets

-0.69

\":

-0.64

 pretext

-0.64

Accessory

-0.64

bum

-0.62

 question

-0.62

 scratch

-0.62

 suggestion

-0.61

anuts

-0.60

 pals

-0.60

POSITIVE LOGITS

ithing

0.76

kefeller

0.76

ĸļ

0.74

hester

0.73

Zone

0.72

ptroller

0.71

heast

0.71

agara

0.71

esan

0.70

umbn

0.68

Activations Density 0.131%

locations

geographical locations and their associated coordinates or identifiers

No Comments

No Known Activations