INDEX

Explanations

phrases related to various categories or types

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

uil

-0.17

 Hind

-0.15

ffen

-0.14

Til

-0.14

 Arte

-0.14

ÅĽnie

-0.13

swick

-0.13

Ih

-0.13

affles

-0.13

POSITIVE LOGITS

forme

0.17

'gc

0.16

istrov

0.15

 Karlov

0.14

remium

0.14

_raise

0.14

usi

0.14

styleType

0.14

ë¹Ħ

0.14

 Garner

0.14

Activations Density 0.006%

phrases related to various categories or types

No Comments

No Known Activations