INDEX

Explanations

phrases related to people and their actions

oai_token-act-pair · gpt-3.5-turbo

the word "and" indicating collaboration or connection between entities

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

uggest

-0.87

=]

-0.74

Recommend

-0.74

cms

-0.74

 WARN

-0.70

venant

-0.70

IST

-0.67

.–

-0.67

amer

-0.67

:(

-0.67

POSITIVE LOGITS

 colleagues

0.90

 others

0.89

 vice

0.88

rogen

0.84

 Sons

0.82

rogens

0.78

 Associates

0.77

 coworkers

0.77

 fellow

0.75

 consequently

0.75

Activations Density 0.330%

phrases related to people and their actions

the word "and" indicating collaboration or connection between entities

No Comments

No Known Activations