INDEX

Explanations

words related to future events or predictions

oai_token-act-pair · gpt-3.5-turbo

the concept of "next" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

lees

-0.89

theless

-0.75

lee

-0.75

xual

-0.75

 Lauder

-0.69

tics

-0.67

then

-0.66

enance

-0.66

balls

-0.65

still

-0.64

POSITIVE LOGITS

 generation

1.12

 installment

1.08

 iteration

1.04

 incarnation

0.94

 step

0.92

decade

0.91

 phase

0.91

 generations

0.89

 chapter

0.89

 occupant

0.88

Activations Density 0.051%

words related to future events or predictions

the concept of "next" in various contexts

No Comments

No Known Activations