INDEX

Explanations

tasks or steps in a set of instructions

oai_token-act-pair · gpt-3.5-turbo

sequences of numbered steps in instructional or procedural texts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

soDeliveryDate

-0.79

 likeness

-0.67

 minded

-0.67

 nineteen

-0.64

athered

-0.61

atorial

-0.61

 passer

-0.60

 estranged

-0.60

 hungry

-0.59

 stranger

-0.58

POSITIVE LOGITS

1.15

:{

1.01

 onwards

0.93

0.92

:]

0.92

 Edit

0.89

 Procedure

0.85

 Conclusion

0.84

.:

0.83

:-

0.83

Activations Density 0.070%

tasks or steps in a set of instructions

sequences of numbered steps in instructional or procedural texts

No Comments

No Known Activations