INDEX

Explanations

specific terms and phrases related to document and file structure, along with certain numerical or code-like elements

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

DSP

-0.15

ÙĪØ§Ø±

-0.15

abric

-0.14

phen

-0.14

laden

-0.14

WithPath

-0.14

ottie

-0.14

arios

-0.14

tru

-0.13

toa

-0.13

POSITIVE LOGITS

æ·»

0.17

okens

0.15

oland

0.15

odb

0.15

çļ®

0.14

BufferData

0.14

nell

0.14

oj

0.14

unker

0.14

literal

0.13

Activations Density 0.010%

specific terms and phrases related to document and file structure, along with certain numerical or code-like elements

No Comments

No Known Activations

specific terms and phrases related to document and file structure, along with certain numerical or code-like elements

No Comments

No Known Activations