INDEX

Explanations

phrases that convey positive evaluation or approval

oai_token-act-pair · gpt-3.5-turbo

sentences that conclude with a period

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

²

-0.69

 conspicuous

-0.67

aband

-0.66

 habit

-0.66

unker

-0.65

aganda

-0.65

ixture

-0.64

isphere

-0.63

 accumulated

-0.63

 staggered

-0.63

POSITIVE LOGITS

 Adds

1.02

 Asked

0.98

<|endoftext|>

0.94

 Ultimately

0.93

 Saying

0.91

 Refer

0.88

 Nevertheless

0.87

 Exactly

0.87

 âĢķ

0.87

 Such

0.87

Activations Density 0.086%

phrases that convey positive evaluation or approval

sentences that conclude with a period

No Comments

No Known Activations