INDEX

Explanations

dates written in the format "Month Dayth"

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, particularly periods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 downright

-0.80

 forgetting

-0.77

 deserved

-0.74

 conceivable

-0.73

 frankly

-0.72

 tremend

-0.71

 slapping

-0.71

 smelling

-0.69

 trusted

-0.69

 unthinkable

-0.69

POSITIVE LOGITS

<|endoftext|>

1.50

 Tickets

1.13

 Coverage

1.08

 Â®

1.01

Its

1.01

 Previous

1.00

 Previously

0.98

 Admission

0.98

 Prior

0.98

 Additionally

0.98

Activations Density 0.321%

dates written in the format "Month Dayth"

punctuation marks, particularly periods

No Comments

No Known Activations