INDEX

Explanations

acronyms, titles, and geographical identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

stad

-0.16

lesc

-0.15

ibern

-0.15

alars

-0.14

utm

-0.14

à¹Īà¸²à¸ĩ

-0.13

kaar

-0.13

_deposit

-0.13

Ð»Ð°ÑĪ

-0.13

POSITIVE LOGITS

Ã£

0.15

CallCheck

0.15

civ

0.14

ÏģÎ¿Ïħ

0.14

idden

0.14

OffsetTable

0.14

oram

0.14

eden

0.14

ullen

0.13

NAS

0.13

Activations Density 0.087%

acronyms, titles, and geographical identifiers

No Comments

No Known Activations