INDEX

Explanations

words related to communication and explanations

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate clarification or informative communication

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

 withdrawn

-0.77

 stunts

-0.76

soever

-0.69

 coerc

-0.69

 interfered

-0.68

 risky

-0.67

yss

-0.67

 gamble

-0.65

 sham

-0.65

 disobedience

-0.65

POSITIVE LOGITS

 concise

1.37

 succinct

1.33

 clarity

1.24

 clarify

1.20

 clearer

1.16

 understanding

1.16

 overview

1.16

 outline

1.15

 understand

1.11

 explaining

1.09

Activations Density 0.413%

words related to communication and explanations

phrases that indicate clarification or informative communication

No Comments

No Known Activations