INDEX

Explanations

phrases that involve evaluation or judgment of individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ss

-0.17

æģĴ

-0.15

iler

-0.15

rist

-0.15

ãĥ«ãĤ¯

-0.14

 Gregory

-0.14

ucher

-0.14

sv

-0.14

abelle

-0.14

ff

-0.13

POSITIVE LOGITS

ìĪ

0.16

éĽĳ

0.15

éĻµ

0.15

coma

0.15

rette

0.15

usercontent

0.15

ÅĻet

0.15

jed

0.15

Ø±ÙĪØª

0.14

reta

0.14

Activations Density 0.287%

phrases that involve evaluation or judgment of individuals or entities

No Comments

No Known Activations