INDEX

Explanations

verbs expressing continuation or persistence

oai_token-act-pair · gpt-3.5-turbo

repeated phrases emphasizing continuity or persistence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

anooga

-0.67

 Explan

-0.64

 Prem

-0.59

posted

-0.59

soDeliveryDate

-0.59

most

-0.59

Runner

-0.58

eport

-0.58

grounds

-0.58

 Surprise

-0.57

POSITIVE LOGITS

 explore

0.94

 enjoy

0.92

ggles

0.91

 earn

0.89

 satisfy

0.89

 strengthen

0.88

 accumulate

0.88

 comply

0.87

 negotiate

0.87

 rely

0.86

Activations Density 0.042%

verbs expressing continuation or persistence

repeated phrases emphasizing continuity or persistence

No Comments

No Known Activations