INDEX

Explanations

the word "last" followed by a number

oai_token-act-pair · gpt-3.5-turbo

instances of the word "last" 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ingen

-0.91

cil

-0.71

nel

-0.67

selves

-0.66

amber

-0.65

ancel

-0.64

hire

-0.64

ourage

-0.62

velt

-0.62

gravity

-0.62

POSITIVE LOGITS

 gasp

1.12

 vest

1.01

 ditch

0.95

 straw

0.93

 rites

0.90

 remnant

0.90

decade

0.88

 bast

0.88

 installment

0.83

 remaining

0.82

Activations Density 0.062%

the word "last" followed by a number

instances of the word "last" in various contexts

No Comments

No Known Activations