INDEX

Explanations

numerical values denoting certainty or probability, particularly when expressed as percentages

oai_token-act-pair · gpt-3.5-turbo

phrases indicating certainty or quantifiable metrics, especially those expressing percentages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

rift

-0.78

vati

-0.73

rang

-0.72

colm

-0.71

olas

-0.70

isen

-0.70

netflix

-0.69

achi

-0.69

akings

-0.68

attled

-0.67

POSITIVE LOGITS

1.21

%"

1.03

%]

0.96

0.95

%;

0.90

=#

0.89

Percent

0.86

0.85

Activations Density 0.027%

numerical values denoting certainty or probability, particularly when expressed as percentages

phrases indicating certainty or quantifiable metrics, especially those expressing percentages

No Comments

No Known Activations