INDEX

Explanations

adjectives related to processes or abstract concepts

oai_token-act-pair · gpt-3.5-turbo

critical phrases related to significant consequences or risks

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

-0.75

,"

-0.72

%,

-0.70

-0.69

!,

-0.69

,'

-0.68

.,

-0.68

.;

-0.68

%;

-0.67

.]

-0.66

POSITIVE LOGITS

etheless

0.90

efully

0.83

 newcom

0.67

 teasp

0.65

sequently

0.65

urther

0.65

ventus

0.64

DragonMagazine

0.61

lly

0.61

PsyNetMessage

0.61

Activations Density 0.952%

adjectives related to processes or abstract concepts

critical phrases related to significant consequences or risks

No Comments

No Known Activations