INDEX

Explanations

words related to serving a specific time period or engagement, such as "stint"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "stint" and related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

hammad

-0.79

ported

-0.72

 Federation

-0.67

 Science

-0.65

health

-0.65

nature

-0.63

cussion

-0.63

âĹ¼

-0.62

classic

-0.62

Quality

-0.61

POSITIVE LOGITS

 stint

1.27

ayson

0.88

 assignments

0.85

 experien

0.83

nell

0.82

 destro

0.79

 abroad

0.77

eatures

0.76

elman

0.75

ings

0.74

Activations Density 0.009%

words related to serving a specific time period or engagement, such as "stint"

occurrences of the word "stint" and related contexts

No Comments

No Known Activations