INDEX

Explanations

phrases related to achieving desired outcomes or results

oai_token-act-pair · gpt-3.5-turbo

terms related to desired outcomes and predetermined conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

vere

-0.74

udi

-0.69

UGH

-0.67

AIDS

-0.65

~~~~~~~~~~~~~~~~

-0.65

 Twain

-0.64

Film

-0.64

 Hasan

-0.63

ologists

-0.63

bane

-0.63

POSITIVE LOGITS

 amount

1.40

 number

1.15

 amounts

1.10

 sized

1.09

 shape

1.08

 outcome

1.08

 quantity

1.07

 values

1.07

 percentage

1.06

 value

1.06

Activations Density 0.245%

phrases related to achieving desired outcomes or results

terms related to desired outcomes and predetermined conditions

No Comments

No Known Activations