INDEX

Explanations

detailed descriptions of actions or processes with a focus on comparisons or specific elements

oai_token-act-pair · gpt-3.5-turbo

prepositions and words indicating relationships or comparisons

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

ities

-0.81

riages

-0.72

 channels

-0.68

 piles

-0.63

inas

-0.63

liners

-0.63

venants

-0.61

chell

-0.61

Â¯Â¯Â¯Â¯Â¯Â¯Â¯Â¯

-0.60

ples

-0.60

POSITIVE LOGITS

 theirs

0.80

 titled

0.74

åĮ

0.70

èĪ

0.68

 behalf

0.66

ordinate

0.64

 Scarlet

0.64

isse

0.62

 Called

0.61

 Named

0.60

Activations Density 0.643%

detailed descriptions of actions or processes with a focus on comparisons or specific elements

prepositions and words indicating relationships or comparisons

No Comments

No Known Activations