INDEX

Explanations

phrases related to beliefs, thoughts, and judgments

oai_token-act-pair · gpt-3.5-turbo

phrases indicating judgments, conclusions, or beliefs about situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Pg

-0.79

vet

-0.71

xtap

-0.69

unal

-0.68

arse

-0.64

ivalent

-0.64

voice

-0.64

izons

-0.64

appropriately

-0.64

depth

-0.64

POSITIVE LOGITS

 THEY

1.09

 they

0.95

 there

0.91

 THERE

0.87

 that

0.79

 nobody

0.76

we

0.74

 thats

0.74

 THAT

0.71

SHE

0.68

Activations Density 0.505%

phrases related to beliefs, thoughts, and judgments

phrases indicating judgments, conclusions, or beliefs about situations

No Comments

No Known Activations