INDEX

Explanations

descriptions of locations and the objects housed within them

oai_token-act-pair · gpt-3.5-turbo

references to historical or iconic objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Cancel

-0.77

 Worse

-0.68

:(

-0.68

 Again

-0.66

Could

-0.65

Would

-0.64

 Delete

-0.63

 Possible

-0.63

pite

-0.63

 Nope

-0.62

POSITIVE LOGITS

 unparalleled

1.06

 unmatched

0.94

 revered

0.92

 annually

0.92

 countless

0.92

 distinctive

0.91

 proudly

0.89

 colorful

0.88

 tirelessly

0.88

quartered

0.87

Activations Density 1.148%

descriptions of locations and the objects housed within them

references to historical or iconic objects

No Comments

No Known Activations