INDEX

Explanations

phrases related to organizations, specifically with a focus on acronyms and classifications

oai_token-act-pair · gpt-3.5-turbo

references to operational protocols or policy frameworks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ãĥīãĥ©ãĤ´ãĥ³

-0.83

taboola

-0.76

perse

-0.74

 Sins

-0.66

Qi

-0.66

brates

-0.64

brate

-0.63

 lashes

-0.63

 Adjusted

-0.62

ãĥĨãĤ£

-0.62

POSITIVE LOGITS

yright

1.12

TION

1.09

rompt

1.08

olicy

1.08

rint

1.07

etition

1.07

rodu

1.02

sych

0.99

inion

0.97

arent

0.96

Activations Density 0.011%

phrases related to organizations, specifically with a focus on acronyms and classifications

references to operational protocols or policy frameworks

No Comments

No Known Activations