INDEX

Explanations

common conjunctions, prepositions, and linking words that connect phrases and ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 architekt

-0.15

mau

-0.15

itat

-0.15

áº«u

-0.15

 Ø§ÙĦÙħÙĩ

-0.14

ingo

-0.14

uv

-0.14

umat

-0.14

undle

-0.14

_HINT

-0.14

POSITIVE LOGITS

caller

0.15

 Liberty

0.15

 showDialog

0.15

 speaker

0.15

iron

0.15

 ÑģÐ¾Ð±Ð¾Ñİ

0.15

stial

0.14

LN

0.14

íĳľ

0.14

coal

0.14

Activations Density 0.005%

common conjunctions, prepositions, and linking words that connect phrases and ideas

No Comments

No Known Activations