INDEX

Explanations

the presence of specific nouns and terms related to museums, social interactions, and measurements of time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

mach

-0.16

ëĤľ

-0.15

ahi

-0.15

osh

-0.15

ci

-0.15

ernal

-0.15

ÑĨÐµÐ½ÑĤÑĢÐ°

-0.14

chio

-0.14

 Maid

-0.14

POSITIVE LOGITS

gos

0.17

ãĥ³ãĥĨãĤ£

0.16

ãĤ°ãĥ©

0.15

ÑĢÐ¾Ð»

0.15

llib

0.14

ogi

0.14

à¸°

0.14

gro

0.14

 Rocky

0.14

ieu

0.14

Activations Density 0.357%

the presence of specific nouns and terms related to museums, social interactions, and measurements of time

No Comments

No Known Activations