INDEX

Explanations

phrases related to following directions or complying with set standards

oai_token-act-pair · gpt-3.5-turbo

instances of the word "by", indicating a focus on agency or attribution in actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

olit

-0.80

ONEY

-0.78

uality

-0.77

ival

-0.77

ppa

-0.76

ussion

-0.76

ashtra

-0.75

iva

-0.75

aughed

-0.74

ovy

-0.73

POSITIVE LOGITS

 virtue

1.07

laws

0.97

 successive

0.80

products

0.77

 statute

0.69

 multiplying

0.67

 predecessors

0.66

 nature

0.66

Tit

0.65

 Hurricane

0.64

Activations Density 0.123%

phrases related to following directions or complying with set standards

instances of the word "by", indicating a focus on agency or attribution in actions

No Comments

No Known Activations