INDEX

Explanations

dates and events from specific time periods

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "last" followed by a time reference

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ordes

-0.82

aband

-0.74

keyes

-0.66

kaya

-0.65

aunders

-0.65

holes

-0.64

cil

-0.63

adle

-0.63

within

-0.62

 arteries

-0.61

POSITIVE LOGITS

 Month

0.93

 month

0.86

 week

0.81

 year

0.78

 Gleaming

0.78

 ditch

0.77

 Update

0.77

 Year

0.74

 night

0.73

 Updated

0.73

Activations Density 0.041%

dates and events from specific time periods

occurrences of the word "last" followed by a time reference

No Comments

No Known Activations