INDEX

Explanations

phrases that express personal inquiries or requests for information

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

ÐµÐ±

-0.14

aption

-0.14

OKIE

-0.14

mux

-0.14

ONSE

-0.14

_clause

-0.14

iller

-0.13

æ£ĭçīĮ

-0.13

&q

-0.13

onium

-0.13

POSITIVE LOGITS

à¥ĩà¤®

0.15

occo

0.14

ihan

0.14

veh

0.14

Projected

0.13

 tomb

0.13

.tem

0.13

ee

0.13

uggy

0.13

Activations Density 0.034%

phrases that express personal inquiries or requests for information

No Comments

No Known Activations