INDEX

Explanations

instances where the verb "were" is used

oai_token-act-pair · gpt-3.5-turbo

instances of the word "were" indicating past actions or events

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

ledge

-0.72

ocracy

-0.72

dom

-0.71

lag

-0.71

town

-0.70

cycle

-0.67

ocratic

-0.66

otic

-0.66

oire

-0.64

orgetown

-0.62

POSITIVE LOGITS

wolves

1.29

wolf

1.04

ĸļ

0.78

 able

0.75

 suppose

0.74

senal

0.74

 supposed

0.71

ateurs

0.70

nt

0.68

 instrumental

0.67

Activations Density 0.175%

instances where the verb "were" is used

instances of the word "were" indicating past actions or events

No Comments

No Known Activations