INDEX

Explanations

questions and prompts in the text

oai_token-act-pair · gpt-3.5-turbo

questions about definitions and explanations of concepts

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

 scarcely

-0.70

."[

-0.67

akable

-0.65

efully

-0.64

,[

-0.63

_>

-0.62

.</

-0.61

thought

-0.60

tur

-0.60

 thereafter

-0.59

POSITIVE LOGITS

?:

1.12

0.91

0.88

Use

0.83

)?

0.81

'?

0.79

 Funds

0.78

.?

0.78

USE

0.77

 Supported

0.77

Activations Density 0.238%

questions and prompts in the text

questions about definitions and explanations of concepts

No Comments

No Known Activations