INDEX

Explanations

phrases related to specific locations or events

oai_token-act-pair · gpt-3.5-turbo

punctuation, specifically periods

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

orical

-0.73

itia

-0.72

omed

-0.72

imates

-0.71

ETF

-0.70

ogen

-0.68

imus

-0.67

oric

-0.65

ather

-0.65

okin

-0.64

POSITIVE LOGITS

 EntityItem

0.73

CLASSIFIED

0.67

 externalToEVAOnly

0.66

 Calais

0.66

avascript

0.64

 Dealer

0.64

DAQ

0.64

taboola

0.62

 citing

0.62

 Bethlehem

0.62

Activations Density 0.033%

phrases related to specific locations or events

punctuation, specifically periods

No Comments

No Known Activations