INDEX

Explanations

nouns and phrases related to human beings

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "being" and humanity

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

 Mines

-0.75

NRS

-0.73

 Passage

-0.67

LCS

-0.65

ories

-0.63

 Fees

-0.63

Els

-0.60

éĹĺ

-0.59

 corrid

-0.58

inka

-0.58

POSITIVE LOGITS

hood

0.95

 endowed

0.88

who

0.84

 inhab

0.79

 alive

0.78

 immersed

0.77

 beings

0.75

 judged

0.75

 lived

0.75

itar

0.73

Activations Density 0.041%

nouns and phrases related to human beings

references to the concept of "being" and humanity

No Comments

No Known Activations