INDEX

Explanations

names or proper nouns

oai_token-act-pair · gpt-3.5-turbo

names of notable people or entities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

»Ĵ

-0.75

å§«

-0.69

 confir

-0.69

iage

-0.69

ctors

-0.67

 srfAttach

-0.67

ãĥł

-0.66

robe

-0.63

notation

-0.62

"],"

-0.61

POSITIVE LOGITS

 equivalent

0.72

 minus

0.71

 Incarn

0.71

 reborn

0.69

 Certified

0.68

actly

0.68

 envy

0.67

 meets

0.66

 Mecca

0.65

 fault

0.65

Activations Density 0.658%

names or proper nouns

names of notable people or entities

No Comments

No Known Activations