INDEX

Explanations

phrases related to reasoning or explanation

oai_token-act-pair · gpt-3.5-turbo

instances of the words "this" and "that" along with their variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Cho

-0.72

venants

-0.70

lets

-0.68

\\\\\\\\

-0.66

leon

-0.66

mare

-0.64

marks

-0.64

fl

-0.63

ãĤ§

-0.63

ãĥ³ãĤ¸

-0.62

POSITIVE LOGITS

 sake

1.67

 purposes

1.38

 purpose

1.07

 reasons

1.02

ummies

1.01

 reason

0.89

 occasion

0.86

 foreseeable

0.85

 particular

0.80

 upcoming

0.78

Activations Density 0.067%

phrases related to reasoning or explanation

instances of the words "this" and "that" along with their variations

No Comments

No Known Activations