INDEX

Explanations

instances of the word "it" and expressions of entertainment-related content

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

bote

-0.19

ucci

-0.17

awa

-0.16

-0.15

_imm

-0.14

iquid

-0.14

nock

-0.14

nergy

-0.14

ergus

-0.14

ãĥĥãĥĹ

-0.14

POSITIVE LOGITS

éĬĢè¡Į

0.16

Nar

0.15

nar

0.15

 Hund

0.14

 Extreme

0.14

yw

0.14

 extreme

0.14

Od

0.13

jez

0.13

auled

0.13

Activations Density 0.047%

instances of the word "it" and expressions of entertainment-related content

No Comments

No Known Activations