INDEX

Explanations

highly specific terms or proper nouns related to character traits, attributes, or identifiers

oai_token-act-pair · gpt-3.5-turbo

references to "character" and its variations in context

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

liga

-0.72

IED

-0.70

 Lars

-0.66

Ùĩ

-0.64

 Jordanian

-0.64

 Reloaded

-0.63

 Norton

-0.63

¿½

-0.63

 Hollande

-0.63

Kop

-0.61

POSITIVE LOGITS

acters

1.38

istics

1.37

istically

1.07

istic

0.91

 charact

0.91

izations

0.87

character

0.78

verage

0.76

Character

0.75

isations

0.75

Activations Density 0.018%

highly specific terms or proper nouns related to character traits, attributes, or identifiers

references to "character" and its variations in context

No Comments

No Known Activations