INDEX

Explanations

specific names or terms related to places and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ertools

-0.17

.nano

-0.16

ÅĻev

-0.15

hton

-0.15

Dre

-0.15

ÑĩÐµÑģÐºÐ¸

-0.14

undler

-0.14

ìĶ

-0.14

eniable

-0.14

 Ø§ÙĦÙħØ®Øª

-0.14

POSITIVE LOGITS

 Cout

0.16

qu

0.15

board

0.15

_closed

0.14

avel

0.14

ìĬ

0.14

orro

0.14

ita

0.14

ear

0.14

apo

0.14

Activations Density 0.022%

specific names or terms related to places and entities

No Comments

No Known Activations