INDEX

Explanations

patterns related to formatted string representations in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ookie

-0.15

à¸£

-0.15

oho

-0.15

å®¶ä¼Ļ

-0.14

omik

-0.14

lam

-0.14

fty

-0.14

éĶ

-0.14

oom

-0.14

yor

-0.13

POSITIVE LOGITS

ë²Ī

0.18

apesh

0.18

oland

0.17

arend

0.15

 Blanch

0.15

çķª

0.14

 Sokol

0.14

bÄĽ

0.14

erton

0.14

elden

0.14

Activations Density 0.022%

patterns related to formatted string representations in code

No Comments

No Known Activations