INDEX

Explanations

names or terms related to specific individuals or locations

oai_token-act-pair · gpt-3.5-turbo

names of prominent individuals, particularly in the context of certain topics or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

pta

-0.77

Fas

-0.74

misc

-0.70

uyomi

-0.69

visor

-0.68

 sleeve

-0.63

 recru

-0.63

 recruitment

-0.63

gression

-0.63

 torches

-0.62

POSITIVE LOGITS

Osw

0.94

ipeg

0.90

 Loch

0.77

ohl

0.77

agher

0.76

uggets

0.76

icz

0.76

sson

0.75

lain

0.72

roth

0.70

Activations Density 0.012%

names or terms related to specific individuals or locations

names of prominent individuals, particularly in the context of certain topics or events

No Comments

No Known Activations