INDEX

Explanations

references to passages of time, especially focusing on years

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "years"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

iqu

-0.76

ople

-0.74

agonist

-0.65

amina

-0.64

iant

-0.63

 Shack

-0.63

 contag

-0.63

apa

-0.62

 Tray

-0.62

aunch

-0.59

POSITIVE LOGITS

ago

1.39

Ago

1.11

long

0.86

manship

0.82

 olds

0.81

mith

0.79

lasting

0.71

 overdue

0.70

quist

0.69

 elapsed

0.69

Activations Density 0.094%

references to passages of time, especially focusing on years

mentions of the word "years"

No Comments

No Known Activations