INDEX

Explanations

keywords related to names, titles, and labels in various contexts

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

iyim

-0.16

arios

-0.15

quate

-0.14

/pm

-0.14

imen

-0.13

anan

-0.13

Dav

-0.13

akter

-0.13

anos

-0.13

edb

-0.13

POSITIVE LOGITS

obot

0.15

oints

0.14

oltip

0.13

stown

0.13

().'/

0.13

zens

0.13

ijken

0.13

isque

0.13

abe

0.13

etta

0.13

Activations Density 0.397%

keywords related to names, titles, and labels in various contexts

No Comments

No Known Activations