INDEX

Explanations

numbers embedded within text strings

oai_token-act-pair · gpt-3.5-turbo

numbers or numerical data points

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wow

-0.68

 credential

-0.61

league

-0.60

ument

-0.60

vous

-0.59

hip

-0.58

sure

-0.56

rica

-0.56

lane

-0.56

 dubious

-0.55

POSITIVE LOGITS

Ô

0.74

--------------------------------------------------------

0.68

 BaseType

0.64

........

0.58

 Ibid

0.58

 =================

0.57

Solution

0.57

Â¯Â¯Â¯Â¯Â¯Â¯Â¯Â¯

0.56

 lineback

0.56

 Tests

0.56

Activations Density 0.039%

numbers embedded within text strings

numbers or numerical data points

No Comments

No Known Activations