INDEX

Explanations

numerical identifiers related to experimental figures, tables, or data points

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ello

-0.17

bak

-0.15

iegel

-0.15

otts

-0.14

ifs

-0.14

orney

-0.14

æĬ

-0.14

ÐµÐ¼Ð¾

-0.14

Ã©ho

-0.14

orks

-0.14

POSITIVE LOGITS

.scalablytyped

0.23

PageIndex

0.18

 Panels

0.17

 Panel

0.15

 shows

0.15

top

0.15

duk

0.15

AllowAnonymous

0.14

shows

0.14

 Taken

0.14

Activations Density 0.052%

numerical identifiers related to experimental figures, tables, or data points

No Comments

No Known Activations