INDEX

Explanations

Ph.D. (Doctor of Philosophy) references in academic and professional contexts

oai_token-act-pair · gpt-3.5-turbo

academic degrees or doctoral titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 chopping

-0.74

 canned

-0.62

zers

-0.61

 temper

-0.61

humans

-0.60

rem

-0.60

 headlines

-0.60

holders

-0.59

quartered

-0.59

 levers

-0.58

POSITIVE LOGITS

Va

0.87

0.86

Ds

0.83

icio

0.78

Skill

0.77

 Degree

0.77

Div

0.76

ĺħ

0.76

owship

0.71

Ops

0.71

Activations Density 0.040%

Ph.D. (Doctor of Philosophy) references in academic and professional contexts

academic degrees or doctoral titles

No Comments

No Known Activations