INDEX

Explanations

phrases that emphasize evaluation or judgment about a performance or experience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

adoo

-0.14

 khÃºc

-0.14

Ð¾Ð¼

-0.14

 ÙħÙĪÙĨ

-0.14

omb

-0.14

REA

-0.14

sez

-0.13

umbo

-0.13

isel

-0.13

undaki

-0.13

POSITIVE LOGITS

ival

0.16

 anything

0.15

 something

0.15

gress

0.14

 fairly

0.14

 exactly

0.14

 prim

0.14

 quite

0.14

arel

0.13

_lifetime

0.13

Activations Density 0.224%

phrases that emphasize evaluation or judgment about a performance or experience

No Comments

No Known Activations