INDEX

Explanations

locations or places, specifically related to cities or towns

oai_token-act-pair · gpt-3.5-turbo

specific place names or locations

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

uce

-0.90

pelling

-0.74

xual

-0.72

odcast

-0.70

hammad

-0.69

ensional

-0.68

ession

-0.67

unct

-0.66

essed

-0.65

irection

-0.64

POSITIVE LOGITS

wich

1.06

izons

0.85

obyl

0.81

 Yankee

0.77

 Mortal

0.76

 Alley

0.74

 Albion

0.73

 Harbor

0.70

swick

0.70

 Joker

0.68

Activations Density 0.031%

locations or places, specifically related to cities or towns

specific place names or locations

No Comments

No Known Activations