INDEX

Explanations

words related to derogatory remarks or negative connotations

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of meaning, specifically through the prefix "dem" and variations thereof

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

 multif

-0.70

ModLoader

-0.66

 lodging

-0.60

 strawberries

-0.59

 Bengal

-0.59

 Annotations

-0.59

 mids

-0.58

 ILCS

-0.57

 fishing

-0.57

 pond

-0.56

POSITIVE LOGITS

ufact

1.07

oppers

0.79

iewicz

0.79

agement

0.74

kas

0.72

ESA

0.70

eanor

0.69

oppable

0.69

rial

0.68

esson

0.68

Activations Density 0.107%

words related to derogatory remarks or negative connotations

words related to the concept of meaning, specifically through the prefix "dem" and variations thereof

No Comments

No Known Activations