INDEX

Explanations

words related to a particular person's name, specifically "Barton"

oai_token-act-pair · gpt-3.5-turbo

references to specific legislative proposals and notable figures involved in them

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

determination

-0.73

hh

-0.73

ebook

-0.70

itability

-0.66

ariat

-0.66

MM

-0.66

egg

-0.66

orses

-0.63

odiac

-0.63

itals

-0.62

POSITIVE LOGITS

 Barton

1.47

bent

0.92

 Springs

0.85

ouf

0.79

pole

0.78

folios

0.77

leigh

0.75

wick

0.75

boro

0.73

 Tiff

0.73

Activations Density 0.010%

words related to a particular person's name, specifically "Barton"

references to specific legislative proposals and notable figures involved in them

No Comments

No Known Activations