INDEX

Explanations

references to specific lengths of time, primarily focusing on the duration of years

oai_token-act-pair · gpt-3.5-turbo

phrases related to durations or lengths of time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ople

-0.80

ibrary

-0.71

ergic

-0.67

liction

-0.63

Leaks

-0.62

 contag

-0.60

anooga

-0.60

gru

-0.59

ash

-0.59

Safe

-0.58

POSITIVE LOGITS

ago

1.47

Ago

1.23

 olds

1.13

manship

0.84

long

0.81

 overdue

0.74

old

0.73

udo

0.73

 elapsed

0.73

 consecut

0.73

Activations Density 0.136%

references to specific lengths of time, primarily focusing on the duration of years

phrases related to durations or lengths of time

No Comments

No Known Activations