INDEX

Explanations

the word "literally" in the text

oai_token-act-pair · gpt-3.5-turbo

the word "literally" used to emphasize statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

icipated

-0.80

lain

-0.78

winner

-0.76

ies

-0.76

eway

-0.75

ramid

-0.73

lings

-0.71

alez

-0.69

resses

-0.69

iers

-0.69

POSITIVE LOGITS

terday

0.73

ãĤ§

0.71

Ń·

0.71

 reinvent

0.68

ãĤ¡

0.68

UNCH

0.66

 torch

0.65

 speaking

0.65

ruciating

0.64

 piss

0.64

Activations Density 0.015%

the word "literally" in the text

the word "literally" used to emphasize statements

No Comments

No Known Activations