INDEX

Explanations

phrases related to titles and designations within a hierarchy

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Downloadha

-0.76

ATURES

-0.72

entimes

-0.69

ooth

-0.69

ãĤ´ãĥ³

-0.66

 accordingly

-0.66

âĢł

-0.66

 Luffy

-0.64

 Canaver

-0.64

rade

-0.63

POSITIVE LOGITS

 aforementioned

1.03

 smallest

1.01

 utmost

0.92

 latter

0.92

 same

0.92

 greatest

0.88

 highest

0.85

 entirety

0.84

 magnitude

0.83

 simplest

0.83

Activations Density 1.580%

phrases related to titles and designations within a hierarchy

occurrences of the word "the" in various contexts

No Comments

No Known Activations