INDEX

Explanations

words related to "New" or "News"

oai_token-act-pair · gpt-3.5-turbo

repeated occurrences of the word "new."

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

 suspic

-0.75

 Reconstruction

-0.59

uate

-0.59

 releg

-0.58

 unarmed

-0.58

 distracted

-0.58

 cort

-0.58

 insensitive

-0.57

 proc

-0.57

 suppressing

-0.57

POSITIVE LOGITS

een

1.21

estern

1.19

riter

1.18

ritten

1.06

esome

1.06

sburg

1.05

esley

1.04

ITNESS

1.03

atcher

1.02

alker

1.00

Activations Density 0.027%

words related to "New" or "News"

repeated occurrences of the word "new."

No Comments

No Known Activations