INDEX

Explanations

words related to medical and chemical terminology

oai_token-act-pair · gpt-3.5-turbo

terms related to logical reasoning and argumentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ikuman

-0.73

 containment

-0.71

 Pigs

-0.68

 largeDownload

-0.68

zona

-0.67

ORED

-0.66

 ellipt

-0.65

cember

-0.65

ModLoader

-0.65

 hover

-0.64

POSITIVE LOGITS

ogyn

0.81

ucker

0.78

andise

0.75

achus

0.74

yll

0.74

agogue

0.74

opol

0.73

urg

0.72

andum

0.72

asury

0.71

Activations Density 0.109%

words related to medical and chemical terminology

terms related to logical reasoning and argumentation

No Comments

No Known Activations