INDEX

Explanations

instances of the word "a" followed by an adjective of various intensity

oai_token-act-pair · gpt-3.5-turbo

phrases that contain the article "a" or "an," indicating a focus on articles in sentence structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Contents

-0.82

Links

-0.77

aura

-0.71

alties

-0.70

insk

-0.69

ells

-0.69

redits

-0.68

upon

-0.67

Events

-0.66

isodes

-0.66

POSITIVE LOGITS

 bunch

1.05

 stranger

1.01

 single

0.97

 politician

0.91

 mere

0.88

 competent

0.87

 woman

0.87

 reputable

0.87

 civilized

0.86

 superpower

0.86

Activations Density 0.434%

instances of the word "a" followed by an adjective of various intensity

phrases that contain the article "a" or "an," indicating a focus on articles in sentence structures

No Comments

No Known Activations