INDEX

Explanations

specific mentions of time in years or numerical time frames

oai_token-act-pair · gpt-3.5-turbo

mention of the word "years" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 contag

-0.72

iqu

-0.72

 Shack

-0.67

ople

-0.67

agonist

-0.65

Indiana

-0.64

amina

-0.63

 stakes

-0.62

aminer

-0.62

 outlet

-0.61

POSITIVE LOGITS

ago

1.37

Ago

1.14

long

0.82

manship

0.78

 olds

0.75

 elapsed

0.73

mith

0.70

 overdue

0.68

ns

0.68

lasting

0.67

Activations Density 0.091%

specific mentions of time in years or numerical time frames

mention of the word "years" and its variations

No Comments

No Known Activations