INDEX

Explanations

words related to philosophical and abstract concepts

oai_token-act-pair · gpt-3.5-turbo

concepts and discussions related to value and understanding

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

HB

-0.59

 Jail

-0.56

lbs

-0.56

 Birmingham

-0.55

 Spotlight

-0.55

Tickets

-0.55

è»

-0.55

FY

-0.55

NES

-0.54

SSL

-0.54

POSITIVE LOGITS

 philosophers

0.86

 subjective

0.81

 epist

0.78

 oneself

0.77

 biologically

0.75

 metaphysical

0.74

 constructs

0.74

 caus

0.73

 logically

0.73

 empirical

0.73

Activations Density 3.342%

words related to philosophical and abstract concepts

concepts and discussions related to value and understanding

No Comments

No Known Activations