INDEX

Explanations

negations and phrases that indicate exceptions or restrictions

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

ADA

-0.16

cken

-0.15

Ø§ÙĪØª

-0.15

ESP

-0.15

inary

-0.14

uren

-0.14

ãĥ¼ãĥ¬

-0.14

 å®ĺ

-0.13

mtree

-0.13

POSITIVE LOGITS

ÏĨÎ±

0.16

Rog

0.15

ereg

0.15

 devis

0.14

aint

0.14

RIPT

0.14

RIX

0.14

epar

0.13

 Nursing

0.13

èªł

0.13

Activations Density 0.004%

negations and phrases that indicate exceptions or restrictions

No Comments

No Known Activations