INDEX

Explanations

phrases related to personal emotions and experiences

oai_token-act-pair · gpt-3.5-turbo

repeated characters or symbols in relation to emotional expressions and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

è£ħ

-0.60

 scattering

-0.60

 juggling

-0.56

 reception

-0.53

çĶ°

-0.53

 spiral

-0.51

 assemb

-0.51

 eleph

-0.51

 dispers

-0.50

 lodging

-0.50

POSITIVE LOGITS

Â¯

0.81

¦

0.79

Â¢

0.75

¬

0.75

£

0.73

¹

0.73

aren

0.73

º

0.72

Ĵ

0.70

acca

0.65

Activations Density 0.419%

phrases related to personal emotions and experiences

repeated characters or symbols in relation to emotional expressions and experiences

No Comments

No Known Activations