INDEX

Explanations

references to a specific person named Trump and issues related to evidence and beliefs about him

oai_token-act-pair · gpt-3.5-turbo

phrases related to belief, opinions, or statements presented as evidence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

versely

-0.49

 contradiction

-0.45

 Arbit

-0.45

ccording

-0.42

)"

-0.40

mosp

-0.40

etheless

-0.40

 misunderstanding

-0.39

urther

-0.39

 correction

-0.38

POSITIVE LOGITS

senal

0.51

è£ħ

0.45

0.44

gins

0.44

raints

0.44

bourg

0.44

geries

0.43

allery

0.43

arsity

0.41

hers

0.40

Activations Density 4.376%

references to a specific person named Trump and issues related to evidence and beliefs about him

phrases related to belief, opinions, or statements presented as evidence

No Comments

No Known Activations