INDEX

Explanations

numerical rankings or positions in lists

oai_token-act-pair · gpt-3.5-turbo

rankings or statistical positions in various sports or competitive contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 revel

-0.60

imb

-0.56

 lessons

-0.56

 coincidence

-0.55

cknow

-0.55

 weaving

-0.53

 Cups

-0.52

cknowled

-0.52

Oo

-0.52

POSITIVE LOGITS

 among

1.08

among

1.05

 nationally

1.00

 amongst

0.98

 overall

0.87

 globally

0.84

 athlet

0.80

 Among

0.78

behind

0.77

 percentile

0.76

Activations Density 0.109%

numerical rankings or positions in lists

rankings or statistical positions in various sports or competitive contexts

No Comments

No Known Activations