INDEX

Explanations

instances of the word "let"

oai_token-act-pair · gpt-3.5-turbo

the word "let" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

oppable

-0.75

 holiest

-0.72

 Languages

-0.72

lihood

-0.69

CLASSIFIED

-0.68

atana

-0.67

nown

-0.66

PLIED

-0.63

 cumbers

-0.61

ãĤ»

-0.61

POSITIVE LOGITS

tered

1.04

icia

0.91

tering

0.83

 slip

0.80

us

0.78

winter

0.72

erest

0.71

eral

0.70

oise

0.69

itia

0.69

Activations Density 0.032%

instances of the word "let"

the word "let" and its variations in different contexts

No Comments

No Known Activations