INDEX

Explanations

concepts related to titles, identities, and choices

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ForegroundColor

-0.15

kick

-0.15

UNT

-0.15

kr

-0.15

 Sirius

-0.15

Salvar

-0.15

ÑĢÐ°Ð´

-0.14

æ°Ĺãģ«åħ¥

-0.14

Worksheet

-0.14

 kick

-0.14

POSITIVE LOGITS

.scalablytyped

0.17

 pent

0.15

 Tamb

0.15

.yy

0.14

Aws

0.14

estroy

0.14

bou

0.14

Tam

0.14

 Econ

0.13

loub

0.13

Activations Density 0.160%

concepts related to titles, identities, and choices

No Comments

No Known Activations