INDEX

Explanations

instances of failure or ineffectiveness in technology or systems

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

ayo

-0.15

clc

-0.14

UpdatedAt

-0.14

izens

-0.14

rops

-0.14

anken

-0.13

 wicked

-0.13

sen

-0.13

ega

-0.13

bun

-0.13

POSITIVE LOGITS

StackNavigator

0.16

eland

0.15

<=$

0.15

ukan

0.15

thead

0.14

 cancelButtonTitle

0.14

;left

0.14

 tubes

0.14

azor

0.14

ccount

0.14

Activations Density 0.008%

instances of failure or ineffectiveness in technology or systems

No Comments

No Known Activations