INDEX

Explanations

proper nouns and specific terms related to different subjects and entities

oai_token-act-pair · gpt-3.5-turbo

names and terms related to cultural or geographic entities

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

ipeg

-0.82

 welf

-0.66

 fodder

-0.65

bucks

-0.63

 appe

-0.61

ogle

-0.61

noon

-0.59

 footing

-0.59

 deductible

-0.59

 diets

-0.59

POSITIVE LOGITS

ioxide

0.77

otide

0.68

Mesh

0.67

Ã±o

0.66

 Alto

0.65

ï¸ı

0.65

oxide

0.65

 Rust

0.64

×Ļ×

0.64

uga

0.62

Activations Density 1.151%

proper nouns and specific terms related to different subjects and entities

names and terms related to cultural or geographic entities

No Comments

No Known Activations