INDEX

Explanations

fragments related to quotations

oai_token-act-pair · gpt-3.5-turbo

rhetorical questions and expressions of opinion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

abouts

-0.94

arij

-0.77

aldi

-0.77

 intended

-0.75

isons

-0.75

oval

-0.74

 herself

-0.73

racuse

-0.70

nesday

-0.70

allas

-0.69

POSITIVE LOGITS

Until

0.95

HAHAHAHA

0.94

Seriously

0.93

Especially

0.86

Anyway

0.85

Lear

0.85

And

0.84

Sure

0.84

Advertisements

0.83

Norm

0.82

Activations Density 0.368%

fragments related to quotations

rhetorical questions and expressions of opinion

No Comments

No Known Activations