INDEX

Explanations

expressions of anticipation and commitment to watching television shows

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

Äĥm

-0.15

 Verde

-0.14

JI

-0.14

.AF

-0.14

ifax

-0.14

 Hari

-0.14

aukee

-0.13

ÐµÐ»Ðµ

-0.13

linger

-0.13

 Weinstein

-0.13

POSITIVE LOGITS

 watch

0.88

watch

0.79

 Watch

0.76

 watching

0.73

Watch

0.72

-watch

0.72

 watches

0.71

 watched

0.69

 WATCH

0.69

.watch

0.68

Activations Density 0.283%

expressions of anticipation and commitment to watching television shows

No Comments

No Known Activations