INDEX

Explanations

phrases that indicate a specific time period or duration

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a time frame or context

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Genie

-0.66

mer

-0.66

 fault

-0.64

bal

-0.63

di

-0.63

por

-0.62

 devil

-0.61

FRE

-0.60

 perpetually

-0.60

 duct

-0.59

POSITIVE LOGITS

¥ŀ

0.98

avorite

0.96

ieth

0.95

ategory

0.94

ruciating

0.93

 Within

0.91

uesday

0.89

htaking

0.88

 Seconds

0.86

ĵĺ

0.85

Activations Density 0.008%

phrases that indicate a specific time period or duration

phrases that indicate a time frame or context

No Comments

No Known Activations