INDEX

Explanations

directives or instructions

oai_token-act-pair · gpt-3.5-turbo

imperative verbs suggesting actions or instructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

cffffcc

-0.77

album

-0.75

operated

-0.66

mith

-0.64

SPONSORED

-0.64

../

-0.63

apo

-0.62

ordered

-0.61

installed

-0.60

\":

-0.59

POSITIVE LOGITS

 Yourself

0.94

 Them

0.86

cknow

0.82

verning

0.79

 yourself

0.77

 Your

0.76

eday

0.75

Mob

0.73

 Advice

0.70

 Appropri

0.70

Activations Density 0.171%

directives or instructions

imperative verbs suggesting actions or instructions

No Comments

No Known Activations