INDEX

Explanations

phrases related to specific past instances or occurrences

oai_token-act-pair · gpt-3.5-turbo

phrases that reference specific locations or contexts within a narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

scl

-0.83

agogue

-0.81

agy

-0.76

 hereby

-0.75

 nonetheless

-0.72

omever

-0.69

 accordingly

-0.68

heit

-0.68

ãĤĬ

-0.67

icas

-0.66

POSITIVE LOGITS

 previous

1.61

 earlier

1.46

 predecessor

1.26

 last

1.24

 predecessors

1.20

 prior

1.19

 preceding

1.19

 inception

1.15

1.14

1.13

Activations Density 0.657%

phrases related to specific past instances or occurrences

phrases that reference specific locations or contexts within a narrative

No Comments

No Known Activations