INDEX

Explanations

instances of the word "Additional"

oai_token-act-pair · gpt-3.5-turbo

the term "additional" and its variations, indicating a focus on supplementary information or resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Alone

-0.74

bane

-0.72

athon

-0.70

bows

-0.69

 Gone

-0.68

ARS

-0.65

lua

-0.65

generic

-0.65

liest

-0.64

esp

-0.64

POSITIVE LOGITS

judicial

0.96

 impetus

0.90

 installments

0.87

 information

0.83

ordinary

0.83

 insight

0.80

 exposure

0.80

 avenues

0.79

 complication

0.79

 layers

0.77

Activations Density 0.020%

instances of the word "Additional"

the term "additional" and its variations, indicating a focus on supplementary information or resources

No Comments

No Known Activations