INDEX

Explanations

punctuation marks indicating the end of statements

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

çĩŁ

-0.15

llen

-0.14

_RX

-0.14

ipple

-0.13

LS

-0.13

èĲ¥

-0.13

rang

-0.13

 Sesso

-0.13

éĻĦ

-0.13

kip

-0.13

POSITIVE LOGITS

ophilia

0.14

Via

0.14

PFN

0.14

Riv

0.14

rias

0.14

icone

0.14

åĪĢ

0.14

osite

0.13

Sur

0.13

omez

0.13

Activations Density 0.000%

punctuation marks indicating the end of statements

No Comments

No Known Activations