INDEX

Explanations

references to coding and technical instructions involving user input and configuration setups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

zek

-0.17

ersh

-0.16

ftware

-0.15

Î®Î½

-0.15

aho

-0.14

ÃŃn

-0.14

acock

-0.14

cete

-0.14

robat

-0.14

oston

-0.14

POSITIVE LOGITS

 enter

0.71

 entered

0.68

 Enter

0.65

Enter

0.64

enter

0.63

 entering

0.63

 enters

0.63

entered

0.56

-enter

0.56

 input

0.55

Activations Density 0.285%

references to coding and technical instructions involving user input and configuration setups

No Comments

No Known Activations