INDEX

Explanations

instances where a large number of people or entities are involved or affected by a particular situation or event

oai_token-act-pair · gpt-3.5-turbo

references to numerous individuals or groups

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

ibal

-0.67

 Presence

-0.64

 opener

-0.63

 largeDownload

-0.62

CHA

-0.62

 presence

-0.61

lim

-0.60

 Loading

-0.60

 Loaded

-0.60

assium

-0.60

POSITIVE LOGITS

 unnamed

0.91

 dozen

0.90

 successive

0.84

 unsuspecting

0.82

 individuals

0.80

 prominent

0.78

 unspecified

0.78

dozen

0.77

 consecutive

0.76

 undisclosed

0.76

Activations Density 0.362%

instances where a large number of people or entities are involved or affected by a particular situation or event

references to numerous individuals or groups

No Comments

No Known Activations