INDEX

Explanations

phrases involving the word "that" followed by a descriptor or action

oai_token-act-pair · gpt-3.5-turbo

phrases containing the word "that" and its relationships to various concepts or situations

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

 citing

-0.64

 foregoing

-0.64

ached

-0.63

 noting

-0.63

 concluding

-0.62

 issuing

-0.61

 fielding

-0.59

 boarding

-0.59

ilot

-0.59

assis

-0.59

POSITIVE LOGITS

 mattered

1.12

 hurts

1.10

 shouldn

1.07

 nobody

1.07

 ought

1.03

 belongs

1.02

 annoy

0.97

 scares

0.97

 horr

0.96

 inspires

0.93

Activations Density 0.205%

phrases involving the word "that" followed by a descriptor or action

phrases containing the word "that" and its relationships to various concepts or situations

No Comments

No Known Activations