INDEX

Explanations

terms related to ambiguity and unclear communication

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

Nano

-0.16

 rÃ³Å¼

-0.15

ltra

-0.15

ÙĨØ¬

-0.15

ral

-0.15

lator

-0.14

 sund

-0.14

amat

-0.14

arket

-0.14

auc

-0.14

POSITIVE LOGITS

Ð¾Ð½

0.15

exist

0.14

ÙĩÙĦ

0.14

 Daemon

0.14

erti

0.14

uae

0.14

asan

0.13

ÑĢÐ¸ÑĦ

0.13

 middle

0.13

Eld

0.13

Activations Density 0.000%

terms related to ambiguity and unclear communication

No Comments

No Known Activations