INDEX

Explanations

technical product names and features

oai_token-act-pair · gpt-3.5-turbo

specific product features and specifications

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 joke

-0.63

 ..."

-0.63

 Tasman

-0.62

 Hilton

-0.61

 nodd

-0.61

 Carlton

-0.60

 Laur

-0.58

attering

-0.58

 Universities

-0.57

************

-0.56

POSITIVE LOGITS

is

1.06

 utilizes

1.02

 consists

1.02

 represents

1.02

 combines

0.97

 provides

0.97

 involves

0.94

 allows

0.93

 manages

0.93

 enables

0.93

Activations Density 0.234%

technical product names and features

specific product features and specifications

No Comments

No Known Activations