INDEX

Explanations

words related to emotions and emotional states, especially focusing on mood

oai_token-act-pair · gpt-3.5-turbo

terms related to mood

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Roe

-0.73

asus

-0.69

 Pipe

-0.67

EX

-0.67

runs

-0.66

wan

-0.64

redit

-0.64

[*]

-0.64

Sources

-0.63

ãĥ¯

-0.63

POSITIVE LOGITS

 mood

3.89

 Mood

2.46

 tone

1.40

 emotions

1.37

 vibe

1.32

 melanch

1.29

 appetite

1.23

 gloom

1.22

 emotion

1.19

 depressive

1.16

Activations Density 0.016%

words related to emotions and emotional states, especially focusing on mood

terms related to mood

No Comments

No Known Activations