INDEX

Explanations

personal pronouns and verbs indicating personal preferences or actions

oai_token-act-pair · gpt-3.5-turbo

occurrences of emotions and personal experiences

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 âĵĺ

-0.68

 Appears

-0.67

 Authors

-0.66

wikipedia

-0.66

iffin

-0.66

NX

-0.65

jri

-0.63

onic

-0.63

ESA

-0.62

bable

-0.62

POSITIVE LOGITS

 outnumbered

0.72

 outwe

0.72

 pree

0.71

 menstru

0.69

 darn

0.68

 cumbers

0.65

 insufficient

0.63

too

0.63

 tremend

0.62

 cheap

0.62

Activations Density 0.452%

personal pronouns and verbs indicating personal preferences or actions

occurrences of emotions and personal experiences

No Comments

No Known Activations