INDEX

Explanations

commands in programming languages, particularly related to indicating errors or actions to perform

oai_token-act-pair · gpt-3.5-turbo

concepts related to relative paths and file system operations

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

iola

-0.93

iol

-0.77

atti

-0.71

abad

-0.69

daq

-0.68

lio

-0.67

ÃŃa

-0.67

aph

-0.66

_>

-0.65

 Gutierrez

-0.64

POSITIVE LOGITS

ggles

0.95

 toggle

0.85

 optional

0.84

 abbre

0.84

 interpol

0.83

 asynchronous

0.77

 horizontal

0.75

trak

0.74

 abbrevi

0.72

 specifies

0.71

Activations Density 0.350%

commands in programming languages, particularly related to indicating errors or actions to perform

concepts related to relative paths and file system operations

No Comments

No Known Activations