INDEX

Explanations

references to cartography and geographical terms

oai_token-act-pair · gpt-3.5-turbo

terms related to various forms of "graphy" or mapping techniques

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

ŃĶ

-0.84

aren

-0.68

base

-0.67

atti

-0.66

inki

-0.65

 loose

-0.65

ership

-0.63

fires

-0.63

tab

-0.63

ention

-0.63

POSITIVE LOGITS

ography

0.88

 Compass

0.79

ographer

0.79

issance

0.76

utra

0.71

ographers

0.70

 Scotia

0.69

ogram

0.69

 silhou

0.68

 Clouds

0.66

Activations Density 0.067%

references to cartography and geographical terms

terms related to various forms of "graphy" or mapping techniques

No Comments

No Known Activations