INDEX

Explanations

prepositions indicating direction or action

oai_token-act-pair · gpt-3.5-turbo

prepositions and phrases indicating relationships or directions

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

olulu

-0.88

daq

-0.78

hirt

-0.73

fab

-0.70

ashtra

-0.68

illac

-0.67

cyclopedia

-0.67

alys

-0.67

nick

-0.66

robe

-0.66

POSITIVE LOGITS

tnc

0.72

PW

0.72

LW

0.71

 onset

0.71

effic

0.71

 âī¤

0.64

GW

0.64

 nonex

0.63

efficiency

0.63

 lieu

0.62

Activations Density 0.476%

prepositions indicating direction or action

prepositions and phrases indicating relationships or directions

No Comments

No Known Activations