INDEX

Explanations

terms related to data analysis

oai_token-act-pair · gpt-3.5-turbo

verbs and nouns related to analysis or evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 paycheck

-0.88

unte

-0.80

 purse

-0.75

 pant

-0.74

 upkeep

-0.74

 carpet

-0.70

 landsc

-0.70

 closet

-0.68

 treadmill

-0.67

 envelope

-0.67

POSITIVE LOGITS

Gaw

0.88

antage

0.86

 Aviv

0.80

REDACTED

0.79

çīĪ

0.78

 Explain

0.77

Obj

0.76

 Canaver

0.73

 BELOW

0.72

 Emin

0.72

Activations Density 0.206%

terms related to data analysis

verbs and nouns related to analysis or evaluation

No Comments

No Known Activations