INDEX

Explanations

phrases related to contrasts or opposing concepts

oai_token-act-pair · gpt-3.5-turbo

phrases related to actions or states of being that imply some form of controversy or contradiction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pires

-0.73

opolis

-0.64

SPA

-0.62

 blames

-0.60

 suffers

-0.60

patient

-0.60

HAS

-0.58

 seeks

-0.58

 Kills

-0.58

atform

-0.58

POSITIVE LOGITS

 themselves

1.35

 respectively

1.22

 apiece

1.10

were

0.98

 were

0.94

 weren

0.90

 aren

0.90

are

0.85

.","

0.79

 expire

0.78

Activations Density 0.799%

phrases related to contrasts or opposing concepts

phrases related to actions or states of being that imply some form of controversy or contradiction

No Comments

No Known Activations