INDEX

Explanations

specific techniques or methods

oai_token-act-pair · gpt-3.5-turbo

references to various techniques and methods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

riel

-0.68

endar

-0.67

onen

-0.67

hoff

-0.64

gur

-0.60

iny

-0.60

à

-0.59

 gloom

-0.59

gin

-0.59

conn

-0.58

POSITIVE LOGITS

ologies

1.13

 techniques

0.93

 pioneered

0.81

 technique

0.78

 tricks

0.77

 methods

0.76

 employed

0.75

 manuals

0.74

 whereby

0.73

 Techniques

0.72

Activations Density 0.055%

specific techniques or methods

references to various techniques and methods

No Comments

No Known Activations