INDEX

Explanations

emphatic statements or expressions

oai_token-act-pair · gpt-3.5-turbo

various punctuation marks and quotations in the text

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

²¾

-0.71

 throne

-0.68

ĵĺ

-0.68

 brill

-0.64

onom

-0.63

egu

-0.62

Ħ¢

-0.62

 reunited

-0.61

itialized

-0.61

 objective

-0.60

POSITIVE LOGITS

 recalls

1.04

 recalled

0.88

 remembers

0.80

 reads

0.76

zn

0.75

according

0.74

 writes

0.74

 recounted

0.73

"],

0.73

said

0.73

Activations Density 0.164%

emphatic statements or expressions

various punctuation marks and quotations in the text

No Comments

No Known Activations