INDEX

Explanations

phrases related to expressions of opinion or disagreement

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the."

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

fried

-0.66

#$

-0.65

ÙĴ

-0.65

esar

-0.65

itia

-0.64

hur

-0.63

 aloud

-0.63

Cho

-0.63

.-

-0.62

ãĥ³

-0.62

POSITIVE LOGITS

 sake

1.65

 purposes

1.35

 foreseeable

1.31

 moment

1.09

 longest

1.08

 unin

1.07

 past

1.05

 meantime

1.05

 duration

1.04

 avoidance

1.00

Activations Density 0.051%

phrases related to expressions of opinion or disagreement

instances of the word "the."

No Comments

No Known Activations