INDEX

Explanations

words related to emotions or moods

oai_token-act-pair · gpt-3.5-turbo

references to emotional states and moods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Universities

-0.74

ces

-0.74

adr

-0.73

xon

-0.72

ced

-0.71

anne

-0.69

Ded

-0.68

ACTED

-0.68

chnology

-0.67

toe

-0.67

POSITIVE LOGITS

 mood

1.11

 Mood

0.96

 contag

0.87

 stabil

0.83

 swings

0.77

achine

0.77

atically

0.74

 comed

0.74

 tone

0.73

0.72

Activations Density 0.010%

words related to emotions or moods

references to emotional states and moods

No Comments

No Known Activations