INDEX

Explanations

references to television shows or movie pilots

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

edith

-0.16

fut

-0.16

-preview

-0.15

Spoiler

-0.15

ivec

-0.15

adder

-0.14

upply

-0.14

è±Ĩ

-0.14

Äı

-0.14

appen

-0.13

POSITIVE LOGITS

EX

0.16

 topl

0.16

PROTO

0.15

_cast

0.15

hel

0.14

pic

0.14

Zem

0.14

inke

0.14

 pilot

0.14

Activations Density 0.043%

references to television shows or movie pilots

No Comments

No Known Activations