INDEX

Explanations

words related to newness or recent events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "newly" 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

trak

-0.86

 Bullets

-0.77

ocene

-0.77

igraph

-0.75

orem

-0.72

rice

-0.70

 Limbaugh

-0.67

 Attribution

-0.67

urgy

-0.67

ashtra

-0.67

POSITIVE LOGITS

ãĤ»

0.99

ãĤ´

0.94

foundland

0.83

 reunited

0.78

 unsus

0.77

bies

0.74

 crowned

0.73

WIND

0.72

 budding

0.71

 discovered

0.70

Activations Density 0.008%

words related to newness or recent events

occurrences of the word "newly" in various contexts

No Comments

No Known Activations