INDEX

Explanations

mentions of news articles or tweets containing the keyword "HT"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "HT" or "ht"

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

ately

-0.69

ptions

-0.68

ators

-0.67

 Samoa

-0.67

ãĥ£

-0.66

ator

-0.66

fman

-0.65

atories

-0.63

ãĥ¼ãĥĨ

-0.63

iott

-0.62

POSITIVE LOGITS

TPS

1.16

ilde

0.97

hered

0.94

ford

0.88

ness

0.84

nikov

0.81

chery

0.81

ulhu

0.81

alin

0.80

alm

0.80

Activations Density 0.032%

mentions of news articles or tweets containing the keyword "HT"

occurrences of the term "HT" or "ht"

No Comments

No Known Activations