INDEX

Explanations

phrases related to well-known or notorious people, places, or events

oai_token-act-pair · gpt-3.5-turbo

descriptors related to notoriety and significance

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

orthy

-0.82

arate

-0.78

Specific

-0.77

ometers

-0.76

cht

-0.76

thia

-0.75

avia

-0.73

AAAA

-0.72

ÄŁ

-0.71

OIL

-0.71

POSITIVE LOGITS

 aftermath

0.76

 industrial

0.72

est

0.70

 nature

0.70

bir

0.70

 onslaught

0.69

liest

0.69

 seas

0.69

 saga

0.68

 intersection

0.68

Activations Density 0.306%

phrases related to well-known or notorious people, places, or events

descriptors related to notoriety and significance

No Comments

No Known Activations