INDEX

Explanations

mentions of reading or recommendations to read

oai_token-act-pair · gpt-3.5-turbo

instances of the word "read."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

pload

-0.82

ascal

-0.79

xon

-0.74

ortality

-0.74

Ã±o

-0.71

unity

-0.71

ugal

-0.70

afort

-0.69

Truth

-0.68

VC

-0.68

POSITIVE LOGITS

 aloud

1.25

just

1.03

 comprehension

0.96

ied

0.85

ying

0.84

dress

0.81

 read

0.78

ahead

0.76

reads

0.74

iness

0.73

Activations Density 0.030%

mentions of reading or recommendations to read

instances of the word "read."

No Comments

No Known Activations