INDEX

Explanations

future predictions or speculations

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate future actions or predictions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.67

arthed

-0.61

Said

-0.61

ZI

-0.61

vati

-0.60

wrote

-0.59

 joking

-0.59

Repl

-0.58

 Recon

-0.57

 Sandwich

-0.57

POSITIVE LOGITS

 undoubtedly

1.44

 doubtless

1.44

 inevitably

1.41

 surely

1.32

 likely

1.26

 probably

1.20

 unavoid

1.17

 certainly

1.13

 require

1.12

be

1.12

Activations Density 0.197%

future predictions or speculations

phrases that indicate future actions or predictions

No Comments

No Known Activations