INDEX

Explanations

phrases related to time or a consecutive order of events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the preposition "in" and related contextual phrases that indicate time or location

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

estamp

-0.68

eps

-0.60

bos

-0.59

bies

-0.59

abe

-0.58

 paran

-0.57

Required

-0.57

 sleeps

-0.57

 asses

-0.57

 enthusi

-0.56

POSITIVE LOGITS

 hindsight

1.11

 subsequent

1.10

 retrospect

0.96

0.90

0.89

0.88

0.86

Activations Density 0.129%

phrases related to time or a consecutive order of events

occurrences of the preposition "in" and related contextual phrases that indicate time or location

No Comments

No Known Activations