INDEX

Explanations

descriptions or discussions about historical events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "events."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

zzi

-0.76

geon

-0.74

sis

-0.67

sburgh

-0.65

 territorial

-0.65

ictionary

-0.64

tesy

-0.63

berman

-0.62

ately

-0.61

arton

-0.61

POSITIVE LOGITS

 events

1.10

hips

0.89

 transpired

0.89

 unfolded

0.89

 unfold

0.86

 Events

0.84

 unfolding

0.80

etter

0.79

hower

0.77

events

0.77

Activations Density 0.018%

descriptions or discussions about historical events

occurrences of the word "events."

No Comments

No Known Activations