INDEX

Explanations

words related to emotions and moods

oai_token-act-pair · gpt-3.5-turbo

references to mood

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

adr

-0.88

ACTED

-0.84

ces

-0.77

anne

-0.77

cing

-0.75

toe

-0.74

advertisement

-0.74

ced

-0.73

Dou

-0.73

pta

-0.69

POSITIVE LOGITS

 mood

1.28

 Mood

1.07

 contag

0.84

achine

0.77

 swings

0.77

 calming

0.77

 conflic

0.75

 stabil

0.75

 tone

0.74

otions

0.73

Activations Density 0.010%

words related to emotions and moods

references to mood

No Comments

No Known Activations