INDEX

Explanations

phrases indicating evaluation or judgment

oai_token-act-pair · gpt-3.5-turbo

phrases related to conditions and limitations involving potential actions or guarantees

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 ï¿½

-0.79

,)

-0.73

.),

-0.72

 ãĢĮ

-0.70

.)

-0.67

.):

-0.66

?),

-0.65

?,

-0.63

 Â§Â§

-0.62

 ----------------

-0.61

POSITIVE LOGITS

%"

1.41

1.31

usterity

1.23

".[

1.19

".

1.13

"?

1.13

'"

1.09

"—

1.06

"!

1.06

"[

1.04

Activations Density 0.353%

phrases indicating evaluation or judgment

phrases related to conditions and limitations involving potential actions or guarantees

No Comments

No Known Activations