INDEX

Explanations

conjunctions and phrases that indicate contrast or exceptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

kiye

-0.17

Overrides

-0.15

itag

-0.15

aupt

-0.15

uci

-0.15

orna

-0.14

cie

-0.14

ngo

-0.14

erto

-0.14

ubl

-0.14

POSITIVE LOGITS

/******/

0.17

chor

0.16

 Prosper

0.15

LookAndFeel

0.14

ovacÃŃ

0.14

mÃ¼r

0.14

 Williamson

0.14

_consts

0.13

 Robbins

0.13

atra

0.13

Activations Density 0.082%

conjunctions and phrases that indicate contrast or exceptions

No Comments

No Known Activations