INDEX

Explanations

phrases related to speed or efficiency

oai_token-act-pair · gpt-3.5-turbo

phrases indicating timeframes or durations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Initialized

-0.70

emy

-0.63

ffen

-0.63

itive

-0.62

idon

-0.62

ãĥī

-0.61

 superiority

-0.61

awan

-0.60

amen

-0.59

Bottom

-0.58

POSITIVE LOGITS

 seconds

1.32

 minutes

1.26

 milliseconds

1.23

seconds

1.18

 weeks

1.11

 hours

1.07

 Seconds

1.07

few

1.05

 days

1.05

 blink

1.03

Activations Density 0.144%

phrases related to speed or efficiency

phrases indicating timeframes or durations

No Comments

No Known Activations