INDEX

Explanations

phrases related to untapped or undeveloped potential

oai_token-act-pair · gpt-3.5-turbo

variations of the suffixes related to taxonomies or classifications

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

aminer

-0.72

rir

-0.70

oother

-0.68

odi

-0.67

utra

-0.67

inus

-0.62

ods

-0.62

cial

-0.61

ickr

-0.59

lev

-0.58

POSITIVE LOGITS

iate

0.80

ness

0.71

aneous

0.71

adoes

0.69

 copies

0.69

kered

0.66

iated

0.66

paren

0.66

 Territories

0.65

ibilities

0.65

Activations Density 0.072%

phrases related to untapped or undeveloped potential

variations of the suffixes related to taxonomies or classifications

No Comments

No Known Activations