INDEX

Explanations

personnel involved in specific events or situations, such as journalists, students, or security officers

oai_token-act-pair · gpt-3.5-turbo

mentions of specific occupations or roles related to various individuals

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

 Pieces

-0.70

 VIDEOS

-0.69

 dots

-0.68

 goblins

-0.68

inters

-0.67

 intervals

-0.66

rams

-0.65

 Hearts

-0.64

Frames

-0.62

 Machines

-0.62

POSITIVE LOGITS

who

1.05

 named

0.97

 whose

0.92

 whom

0.82

alyst

0.81

 classmate

0.80

 nicknamed

0.79

 staffer

0.79

whose

0.77

 specializing

0.77

Activations Density 0.288%

personnel involved in specific events or situations, such as journalists, students, or security officers

mentions of specific occupations or roles related to various individuals

No Comments

No Known Activations