INDEX

Explanations

occurrences of specific names or terms related to a particular location

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

ãĤ¤ãĥ¤

-0.17

idan

-0.16

ationale

-0.16

criptor

-0.15

oft

-0.15

dni

-0.15

htable

-0.14

 ↵↵

-0.14

upiter

-0.14

ador

-0.14

POSITIVE LOGITS

yssey

0.33

essa

0.26

ious

0.19

ys

0.19

ÑıÐ³

0.19

isha

0.17

Od

0.17

gaard

0.16

ette

0.16

IOUS

0.16

Activations Density 0.014%

occurrences of specific names or terms related to a particular location

No Comments

No Known Activations