INDEX

Explanations

words and phrases indicating leadership, organization, and roles in educational or professional contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 Richardson

-0.15

unken

-0.15

jak

-0.15

 McLaren

-0.14

ERIC

-0.14

ãĥªãĥ³ãĤ°

-0.14

ying

-0.14

 Akron

-0.14

ãģıãģ¨

-0.14

dess

-0.13

POSITIVE LOGITS

isol

0.14

Äįin

0.14

egt

0.14

ByUsername

0.14

.AF

0.14

ottes

0.14

PIT

0.14

ãĤ²

0.14

 Cheat

0.14

æĻ¶

0.14

Activations Density 0.005%

words and phrases indicating leadership, organization, and roles in educational or professional contexts

No Comments

No Known Activations