INDEX

Explanations

numbers mentioned in financial or quantitative contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to counting, totals, and sums

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 behaved

-0.73

erate

-0.72

 conditioned

-0.70

scape

-0.65

hov

-0.65

croft

-0.63

oard

-0.62

 friendly

-0.62

 relationship

-0.61

 convict

-0.60

POSITIVE LOGITS

 totals

0.80

uania

0.78

orical

0.78

urgical

0.78

oire

0.76

ories

0.73

tot

0.72

izational

0.71

amaz

0.69

 Estimated

0.68

Activations Density 0.045%

numbers mentioned in financial or quantitative contexts

terms related to counting, totals, and sums

No Comments

No Known Activations