INDEX

Explanations

technical terms related to usage or application

oai_token-act-pair · gpt-3.5-turbo

actions and concepts related to the manipulation and use of information or technology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ItemTracker

-0.76

?)

-0.72

soType

-0.72

unny

-0.67

uyomi

-0.65

cpp

-0.65

 Unsure

-0.64

 convol

-0.63

ãĤ´ãĥ³

-0.62

ogun

-0.62

POSITIVE LOGITS

uate

0.93

 purposes

0.87

 certain

0.73

 various

0.67

enance

0.67

 efficiently

0.66

 plur

0.65

 unsuspecting

0.65

atively

0.65

 subsistence

0.63

Activations Density 0.269%

technical terms related to usage or application

actions and concepts related to the manipulation and use of information or technology

No Comments

No Known Activations