INDEX

Explanations

quantities, counts, and references to significant numbers or measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

Ð½ÐµÑĢÐ³

-0.17

-0.16

away

-0.15

regor

-0.14

plo

-0.14

 lest

-0.14

unken

-0.14

 æ¥

-0.14

all

-0.14

POSITIVE LOGITS

leyin

0.17

toFloat

0.16

ÙİØ¹

0.15

fallback

0.15

elan

0.15

ofile

0.15

PSA

0.14

Br

0.14

dden

0.13

ogenic

0.13

Activations Density 0.003%

quantities, counts, and references to significant numbers or measurements

No Comments

No Known Activations