INDEX

Explanations

references to alcoholic beverages, particularly margaritas

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "itas" 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

itary

-0.84

olicy

-0.70

ãĥķãĤ©

-0.68

 Lena

-0.64

alis

-0.60

ourt

-0.60

 annual

-0.59

 internally

-0.58

 estim

-0.58

 diagn

-0.57

POSITIVE LOGITS

itas

2.44

 kindred

2.36

Muslim

1.25

rawdownloadcloneembedreportprint

1.17

pathy

1.09

SPONSORED

0.97

thus

0.91

scient

0.88

chini

0.85

tnc

0.85

Activations Density 0.033%

references to alcoholic beverages, particularly margaritas

occurrences of the word "itas" in various contexts

No Known Activations