INDEX

Explanations

references to actions and states that indicate existence or ongoing activities

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

terms related to technical specifications or systems

oai_token-act-pair · gpt-3.5-turbo

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

"—

-0.78

)—

-0.74

".[

-0.69

âĹ¼

-0.65

 ÂŃ

-0.63

igious

-0.63

)].

-0.62

Â¶

-0.62

 Diablo

-0.62

 Hearthstone

-0.62

POSITIVE LOGITS

cknow

0.73

 MUST

0.71

cknowled

0.70

 didnt

0.66

 dont

0.63

 beware

0.61

DON

0.57

[-

0.56

Eg

0.55

\'

0.55

Activations Density 0.600%

references to actions and states that indicate existence or ongoing activities

terms related to technical specifications or systems

No Comments

No Known Activations