INDEX

Explanations

scientific terms and references

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Scientific" and related variants

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

owitz

-0.79

cember

-0.79

haul

-0.79

leans

-0.76

choes

-0.75

cles

-0.75

itsch

-0.75

stan

-0.74

gebra

-0.74

vich

-0.74

POSITIVE LOGITS

 Scientific

0.88

ivity

0.80

ally

0.78

 Printed

0.77

 Yankee

0.72

erella

0.72

 Pilgrim

0.72

McA

0.70

unia

0.70

ified

0.69

Activations Density 0.043%

scientific terms and references

occurrences of the word "Scientific" and related variants

No Comments

No Known Activations