INDEX

Explanations

phrases related to contrasts or exceptions

oai_token-act-pair · gpt-3.5-turbo

negations and qualifiers in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

'/

-0.73

 reinforcement

-0.68

 tremend

-0.67

 chill

-0.66

 respons

-0.65

 encl

-0.63

 mature

-0.63

 subp

-0.62

 mans

-0.61

 casing

-0.61

POSITIVE LOGITS

etheless

1.40

theless

1.36

odore

1.18

gotten

1.05

terday

1.02

foundland

1.02

withstanding

1.00

tenance

0.96

ories

0.93

itage

0.92

Activations Density 0.089%

phrases related to contrasts or exceptions

negations and qualifiers in various contexts

No Comments

No Known Activations