INDEX

Explanations

specific names or terms such as "Jude", "Luther", "Mountain", and "Maple"

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

isure

-0.69

phy

-0.64

 leveled

-0.63

bish

-0.63

 Tanz

-0.61

bars

-0.61

 Atomic

-0.59

Pwr

-0.58

 Clockwork

-0.57

 sided

-0.57

POSITIVE LOGITS

iper

0.78

ranch

0.77

jam

0.72

uncture

0.72

hog

0.67

orsche

0.66

oint

0.66

oiler

0.66

umin

0.66

onday

0.66

Activations Density 0.001%

specific names or terms such as "Jude", "Luther", "Mountain", and "Maple"

proper nouns, particularly names and titles

No Comments

No Known Activations