INDEX

Explanations

phrases related to statements or opinions made by specific individuals

oai_token-act-pair · gpt-3.5-turbo

the word "on."

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

AMA

-0.77

ëĭ

-0.73

INESS

-0.71

RL

-0.68

reference

-0.67

orically

-0.66

ACT

-0.65

Mini

-0.63

abo

-0.62

-0.61

POSITIVE LOGITS

 behalf

1.51

etime

1.11

 steroids

1.10

shore

0.99

 occasion

0.94

slaught

0.90

etheless

0.90

eness

0.88

 weekends

0.87

eworld

0.83

Activations Density 0.193%

phrases related to statements or opinions made by specific individuals

the word "on."

No Comments

No Known Activations