INDEX

Explanations

requests and their outcomes, such as being approved, denied, or rejected

oai_token-act-pair · gpt-3.5-turbo

requests and their statuses related to approvals or rejections

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

 inhabit

-0.71

auts

-0.70

 havoc

-0.70

 Spin

-0.70

Variable

-0.69

revolution

-0.66

 Teach

-0.66

Learning

-0.65

 Surviv

-0.65

schild

-0.65

POSITIVE LOGITS

 requesting

1.35

request

1.33

 requests

1.31

 request

1.24

 Request

1.23

Request

1.21

 requested

1.20

 rejected

1.16

requ

1.16

 granted

1.15

Activations Density 0.516%

requests and their outcomes, such as being approved, denied, or rejected

requests and their statuses related to approvals or rejections

No Comments

No Known Activations