INDEX

Explanations

words related to legal matters and medical conditions

oai_token-act-pair · gpt-3.5-turbo

references and terminology related to gender and social roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Kan

-1.01

Kap

-0.99

Gan

-0.92

 Kahn

-0.87

 Mull

-0.87

 Joan

-0.86

 Stefan

-0.86

UX

-0.85

 Khan

-0.84

 Guth

-0.83

POSITIVE LOGITS

ears

1.14

ĵ

0.93

paralle

0.93

orse

0.91

orses

0.87

ear

0.87

oren

0.86

inder

0.86

ory

0.85

eryl

0.85

Activations Density 0.413%

words related to legal matters and medical conditions

references and terminology related to gender and social roles

No Comments

No Known Activations