INDEX

Explanations

comma-separated phrases that emphasize a point or introduce additional details

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "and" in various contexts

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

Zone

-0.71

ELL

-0.65

 guiActiveUnfocused

-0.65

icas

-0.61

utable

-0.61

="#

-0.61

ORN

-0.60

estern

-0.59

olves

-0.59

skirts

-0.57

POSITIVE LOGITS

 consequently

1.23

 importantly

1.16

 moreover

1.15

cru

1.11

 furthermore

1.09

 optionally

1.04

 secondly

1.02

 therefore

1.01

 dare

0.99

 frankly

0.98

Activations Density 0.041%

comma-separated phrases that emphasize a point or introduce additional details

occurrences of the word "and" in various contexts

No Comments

No Known Activations