INDEX

Explanations

phrases indicating a contrasting or explanatory remark

oai_token-act-pair · gpt-3.5-turbo

symbols and special characters that convey emphasis or pauses in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Slug

-0.70

eca

-0.67

 Borders

-0.63

orts

-0.63

 Paso

-0.62

Mob

-0.58

nered

-0.57

 Rhod

-0.56

Mob

-0.56

ORTS

-0.56

POSITIVE LOGITS

 gasp

1.29

 spoiler

1.02

perhaps

1.02

albeit

1.00

 dare

0.99

again

0.98

quite

0.98

despite

0.95

given

0.93

fortunately

0.92

Activations Density 0.122%

phrases indicating a contrasting or explanatory remark

symbols and special characters that convey emphasis or pauses in the text

No Comments

No Known Activations