INDEX

Explanations

occurrences of names and references in media-related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

-0.15

nb

-0.15

-0.14

 Giang

-0.14

etta

-0.14

onya

-0.14

atk

-0.14

anken

-0.14

cles

-0.13

POSITIVE LOGITS

)./

0.15

Natural

0.15

 Natural

0.15

 kino

0.14

uations

0.14

 mest

0.14

indir

0.14

Writes

0.14

GNU

0.14

Activations Density 0.000%

occurrences of names and references in media-related contexts

No Comments

No Known Activations

occurrences of names and references in media-related contexts

No Comments

No Known Activations