INDEX

Explanations

words related to capabilities or characteristics that can be quantified or assessed numerically

oai_token-act-pair · gpt-3.5-turbo

terms related to various types of risks and evaluations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

initions

-0.75

AMES

-0.70

iage

-0.67

HOME

-0.65

UNE

-0.63

 constitution

-0.62

insula

-0.62

 Rasmussen

-0.62

 DeVos

-0.61

ibraries

-0.60

POSITIVE LOGITS

fully

1.11

worthy

1.07

arily

1.01

ishly

1.00

lessly

0.99

ivated

0.94

able

0.92

ivating

0.91

isible

0.91

ueless

0.91

Activations Density 0.389%

words related to capabilities or characteristics that can be quantified or assessed numerically

terms related to various types of risks and evaluations in different contexts

No Comments

No Known Activations