INDEX

Explanations

information that is kept hidden or not revealed

oai_token-act-pair · gpt-3.5-turbo

terms related to undisclosed or unspecified amounts or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

alog

-0.84

ickr

-0.81

ingo

-0.79

=-=-=-=-=-=-=-=-

-0.78

Ont

-0.72

oise

-0.70

ynthesis

-0.70

acco

-0.69

hedral

-0.69

aneers

-0.67

POSITIVE LOGITS

 undisclosed

1.03

 unspecified

0.85

 amounts

0.82

theless

0.76

worldly

0.73

etheless

0.73

 amount

0.73

TBA

0.70

 unnamed

0.70

ties

0.69

Activations Density 0.015%

information that is kept hidden or not revealed

terms related to undisclosed or unspecified amounts or conditions

No Comments

No Known Activations