INDEX

Explanations

significant numerical values or mentions of specific identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

akis

-0.18

marsh

-0.17

.persist

-0.16

oli

-0.15

ÑĤÐ¾Ñĩ

-0.15

ÙħÙĪÙĦ

-0.15

ãĥ³ãĥĳ

-0.14

èªĮ

-0.14

oubles

-0.14

UES

-0.14

POSITIVE LOGITS

yms

0.18

Î³Î¼Î±

0.15

avit

0.15

ells

0.15

egie

0.15

 Epoch

0.14

 Turtle

0.14

asa

0.14

McL

0.14

Activations Density 0.007%

significant numerical values or mentions of specific identifiers

No Comments

No Known Activations