INDEX

Explanations

phrases that emphasize the possessive form of "is" and expressions of affiliation or identity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

 Interr

-0.16

enth

-0.16

åģ

-0.15

Nab

-0.14

_firestore

-0.14

å½

-0.14

uter

-0.13

ÙĤØ¨

-0.13

ial

-0.13

POSITIVE LOGITS

 official

0.23

 Official

0.20

Official

0.20

Alive

0.18

alsy

0.17

 finally

0.16

 been

0.16

finally

0.16

official

0.16

istrovstvÃŃ

0.16

Activations Density 0.053%

phrases that emphasize the possessive form of "is" and expressions of affiliation or identity

No Comments

No Known Activations