INDEX

Explanations

phrases related to comparison or evaluation

oai_token-act-pair · gpt-3.5-turbo

references to relationships or comparisons in terms of various subjects

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

burgh

-0.79

iologist

-0.71

falls

-0.69

pots

-0.68

redited

-0.67

Card

-0.66

enter

-0.64

stars

-0.64

iard

-0.63

izens

-0.63

POSITIVE LOGITS

 sheer

1.03

 chronological

0.83

 geography

0.76

 pure

0.75

 legality

0.75

 geographical

0.73

 fairness

0.73

 keeping

0.72

 proximity

0.72

 improving

0.72

Activations Density 0.045%

phrases related to comparison or evaluation

references to relationships or comparisons in terms of various subjects

No Comments

No Known Activations