INDEX

Explanations

references to specific individuals or events within a narrative context

oai_token-act-pair · gpt-3.5-turbo

specific references to individuals and their associated stories or narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

Downloadha

-0.99

osponsors

-0.89

phabet

-0.89

icons

-0.86

 Thumbnails

-0.76

ledge

-0.75

flags

-0.75

tools

-0.75

emies

-0.74

tails

-0.74

POSITIVE LOGITS

 incident

1.60

 anecdote

1.52

 case

1.29

 instance

1.26

 encounter

1.23

 episode

1.21

 occurrence

1.21

 phenomenon

1.21

 occasion

1.17

 circumstance

1.16

Activations Density 0.375%

references to specific individuals or events within a narrative context

specific references to individuals and their associated stories or narratives

No Comments

No Known Activations