INDEX

Explanations

elements associated with historical exhibitions and educational displays

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

éĽĨ

-0.14

unk

-0.14

hug

-0.14

annon

-0.14

rupa

-0.13

 insulated

-0.13

wat

-0.13

ungi

-0.13

ãĥ¼ãĤ¿

-0.13

ael

-0.13

POSITIVE LOGITS

odate

0.17

_collections

0.15

Ø¨ÙĪØ¯

0.15

 Sinn

0.14

 CreateMap

0.14

 Lehr

0.14

 sacked

0.14

 Objects

0.13

 cure

0.13

Tenant

0.13

Activations Density 0.165%

elements associated with historical exhibitions and educational displays

No Comments

No Known Activations