INDEX

Explanations

references to governmental or institutional entities and their actions

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

phrases related to causes or explanations

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

thood

-0.91

heit

-0.89

strate

-0.79

ature

-0.78

etsy

-0.73

ifies

-0.71

vier

-0.70

ivo

-0.69

ibi

-0.69

jer

-0.68

POSITIVE LOGITS

 sheer

1.14

 aforementioned

1.12

 fact

1.10

 absence

1.02

 influx

1.02

 emergence

1.02

 latter

1.01

 likes

1.01

 presence

1.00

 availability

0.99

Activations Density 0.333%

references to governmental or institutional entities and their actions

phrases related to causes or explanations

No Comments

No Known Activations