INDEX

Explanations

sequences of alphanumeric characters, possibly indicating codes or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.cc

-0.16

dda

-0.15

tere

-0.15

dba

-0.14

 Îī

-0.14

claimer

-0.13

dea

-0.13

dd

-0.13

§è¡Į

-0.13

stuff

-0.13

POSITIVE LOGITS

Uvs

0.18

abwe

0.17

uros

0.16

QueryParam

0.15

 Rounds

0.15

Î»Î»Î·Î½

0.15

áº£o

0.15

ysize

0.15

-Ta

0.15

rating

0.14

Activations Density 0.008%

sequences of alphanumeric characters, possibly indicating codes or identifiers

No Comments

No Known Activations