INDEX

Explanations

phrases and concepts related to human experience and identity

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

ysl

-0.17

vez

-0.16

zn

-0.15

ys

-0.15

lew

-0.15

esso

-0.15

Ø³Ø¯

-0.14

 positional

-0.14

èĨ

-0.14

asd

-0.14

POSITIVE LOGITS

ummings

0.16

agna

0.15

acci

0.15

ucchini

0.14

itoris

0.14

Layers

0.14

tron

0.14

sold

0.14

ÑĤÐ°Ð»

0.14

 whom

0.14

Activations Density 0.236%

phrases and concepts related to human experience and identity

No Comments

No Known Activations