INDEX

Explanations

specific names or references to popular culture elements and media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

((↵

-0.15

lap

-0.14

Ð²Ð¾Ð´

-0.13

$MESS

-0.13

à¥Ĥà¤Ł

-0.13

æī±

-0.13

wort

-0.13

à¥ĭà¤¦

-0.13

ors

-0.13

 Vapor

-0.13

POSITIVE LOGITS

efe

0.15

nt

0.15

Ow

0.14

ele

0.14

 Tango

0.13

ATCH

0.13

alles

0.13

 Gemini

0.13

 inplace

0.13

ango

0.13

Activations Density 0.074%

specific names or references to popular culture elements and media

No Comments

No Known Activations