INDEX

Explanations

phrases related to instructions or requirements in a digital context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 sore

-0.17

ajan

-0.15

 prefer

-0.15

fat

-0.14

Wo

-0.14

ÙıØ³

-0.14

ector

-0.14

-alist

-0.13

 mand

-0.13

 ÑĥÐ´Ð°

-0.13

POSITIVE LOGITS

 blah

0.20

ayo

0.15

illis

0.15

elper

0.15

ÏĢÎŃ

0.14

 Ð´Ð²Ð¾ÑĢ

0.14

æĤ¨çļĦ

0.14

htable

0.13

paring

0.13

Activations Density 0.042%

phrases related to instructions or requirements in a digital context

No Comments

No Known Activations