INDEX

Explanations

phrases related to chemical levels and contamination levels

oai_token-act-pair · gpt-3.5-turbo

numerical comparisons concerning safety limits or health standards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ongs

-0.68

OUP

-0.68

behind

-0.63

ument

-0.62

ong

-0.61

vind

-0.61

 Matters

-0.60

 Networks

-0.60

 Projects

-0.59

PAC

-0.59

POSITIVE LOGITS

 threshold

1.57

 limit

1.48

 thresholds

1.41

 maximum

1.38

 guideline

1.34

 average

1.33

maximum

1.31

 cutoff

1.31

 minimum

1.31

 limits

1.25

Activations Density 0.281%

phrases related to chemical levels and contamination levels

numerical comparisons concerning safety limits or health standards

No Comments

No Known Activations