INDEX

Explanations

specific references to particular objects or entities, especially when they are prefixed with "the"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

thood

-0.54

 without

-0.51

tackle

-0.51

 besides

-0.48

 with

-0.48

-0.47

 because

-0.47

SPONSORED

-0.47

âĢł

-0.47

 âĢº

-0.46

POSITIVE LOGITS

 latter

1.01

 aforementioned

0.98

 slightest

0.94

 same

0.93

 entirety

0.93

 remainder

0.91

 entire

0.89

 smallest

0.86

 latest

0.84

 simplest

0.82

Activations Density 1.463%

specific references to particular objects or entities, especially when they are prefixed with "the"

instances of the word "the" and its variations

No Comments

No Known Activations