INDEX

Explanations

instances of the word "Filed" to categorize content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

swire

-0.15

nez

-0.14

amil

-0.14

patial

-0.14

oming

-0.14

_epi

-0.13

_mE

-0.13

Ã©ru

-0.13

orz

-0.13

lox

-0.13

POSITIVE LOGITS

chten

0.17

esh

0.16

ÄŁ

0.16

elik

0.15

bash

0.15

obot

0.14

elic

0.14

çĦ¼

0.14

Ð¾Ð¿Ð°Ñģ

0.14

licken

0.14

Activations Density 0.003%

instances of the word "Filed" to categorize content

No Comments

No Known Activations