INDEX

Explanations

information about locations and historical significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

illas

-0.16

 ActionTypes

-0.15

fers

-0.14

 klim

-0.13

 Mits

-0.13

Von

-0.13

arden

-0.13

igos

-0.13

ago

-0.13

vent

-0.13

POSITIVE LOGITS

.Butter

0.15

 ãĤ°

0.14

ANGER

0.14

cest

0.14

("-

0.14

kara

0.14

áº©m

0.14

è¾°

0.13

.setViewport

0.13

.preview

0.13

Activations Density 0.210%

information about locations and historical significance

No Comments

No Known Activations