INDEX

Explanations

time-related phrases like "week after", "day before", or "months after"

oai_token-act-pair · gpt-3.5-turbo

temporal phrases indicating time intervals between events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

itement

-0.70

oyal

-0.67

osterone

-0.66

existent

-0.64

ruciating

-0.63

anooga

-0.62

iveness

-0.59

rendered

-0.59

bsite

-0.59

 nonexistent

-0.58

POSITIVE LOGITS

Sin

0.79

furt

0.78

 reinforcements

0.72

 tantal

0.71

lins

0.69

 Canaver

0.67

 heels

0.67

 handy

0.65

nings

0.65

println

0.64

Activations Density 0.133%

time-related phrases like "week after", "day before", or "months after"

temporal phrases indicating time intervals between events

No Comments

No Known Activations