INDEX

Explanations

keywords from scientific figures and data analysis

oai_token-act-pair · gpt-3.5-turbo

references to figures and illustrations in a scientific context

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

minist

-0.78

 polic

-0.76

ishops

-0.70

itability

-0.69

rights

-0.67

istant

-0.67

role

-0.66

raught

-0.65

itiz

-0.65

enez

-0.65

POSITIVE LOGITS

Figure

1.20

 Figure

0.89

Fig

0.85

Table

0.83

 Drawing

0.81

 illustrating

0.78

 depicts

0.78

Ī

0.77

 Exhibit

0.75

FIG

0.74

Activations Density 0.020%

keywords from scientific figures and data analysis

references to figures and illustrations in a scientific context

No Comments

No Known Activations