INDEX

Explanations

references to rankings or numerical identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

å½¹

-0.16

PRINTF

-0.15

ese

-0.15

razil

-0.15

EEP

-0.15

.BLL

-0.15

à¤°à¤£

-0.14

RelativeTo

-0.13

.mit

-0.13

anter

-0.13

POSITIVE LOGITS

uddled

0.14

inds

0.14

Ð¸ÑĢÑĥ

0.14

-notch

0.14

ivot

0.13

 undefeated

0.13

-selling

0.13

portun

0.13

Ã¤lt

0.13

/top

0.13

Activations Density 0.015%

references to rankings or numerical identifiers

No Comments

No Known Activations