INDEX

Explanations

information related to computer programming and technical issues

oai_token-act-pair · gpt-3.5-turbo

terms related to investigations and legal matters

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

 sane

-0.67

 thoughtful

-0.58

 recognizes

-0.57

 wise

-0.57

 civilized

-0.57

 careful

-0.56

ciplinary

-0.55

raged

-0.54

iless

-0.54

 mindful

-0.53

POSITIVE LOGITS

.''.

1.06

.</

0.97

.''

0.95

'.

0.91

''.

0.91

$.

0.89

`.

0.89

.'

0.87

%.

0.86

".

0.84

Activations Density 1.310%

information related to computer programming and technical issues

terms related to investigations and legal matters

No Comments

No Known Activations