INDEX

Explanations

phrases where someone is adding information

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the act of making contributions or additions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ograms

-0.72

bin

-0.69

 ..........

-0.66

 writ

-0.65

debian

-0.63

walk

-0.62

bia

-0.62

 Dreams

-0.61

 Zurich

-0.61

ous

-0.61

POSITIVE LOGITS

itionally

0.87

eki

0.85

endum

0.84

adding

0.83

 ingred

0.82

added

0.77

 insult

0.77

 inval

0.76

eston

0.76

eele

0.74

Activations Density 0.020%

phrases where someone is adding information

phrases indicating the act of making contributions or additions

No Comments

No Known Activations