INDEX

Explanations

numerically indexed data or elements in a structured format

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

enton

-0.15

#@

-0.14

 bail

-0.14

isko

-0.14

est

-0.14

pawn

-0.14

 narrow

-0.14

 rope

-0.14

erton

-0.14

POSITIVE LOGITS

Ð»ÑĥÐ³Ð¸

0.19

unky

0.16

erras

0.15

Ð»ÑĥÐ³

0.14

Raise

0.14

iversary

0.14

eneric

0.13

rof

0.13

ugi

0.13

ispers

0.13

Activations Density 0.012%

numerically indexed data or elements in a structured format

No Comments

No Known Activations