INDEX

Explanations

statements or opinions expressed with certainty

oai_token-act-pair · gpt-3.5-turbo

keywords related to authority and governance

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

 Reef

-0.54

ciplinary

-0.47

 Rasmussen

-0.45

hess

-0.45

ASA

-0.43

 Marshal

-0.41

Kro

-0.41

bard

-0.41

 Lenn

-0.40

 âĨĳ

-0.40

POSITIVE LOGITS

\)

0.56

usterity

0.53

Â»

0.52

ãĥķãĤ¡

0.50

vernment

0.48

-)

0.47

oir

0.46

_.

0.46

`,

0.45

[/

0.44

Activations Density 0.822%

statements or opinions expressed with certainty

keywords related to authority and governance

No Comments

No Known Activations