INDEX

Explanations

phrases that express evaluation or judgement of a subject

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

vlc

-0.17

dent

-0.16

cela

-0.15

 å¾Ĵ

-0.15

htag

-0.15

ienes

-0.14

 Ø¢Ø¯Ùħ

-0.14

hv

-0.14

ertino

-0.14

ãĤ¯ãĥŃ

-0.14

POSITIVE LOGITS

less

0.17

los

0.15

ank

0.15

Mev

0.15

cribes

0.14

ENC

0.14

ìĦŃ

0.14

(calc

0.14

wis

0.14

0.13

Activations Density 0.764%

phrases that express evaluation or judgement of a subject

No Comments

No Known Activations