INDEX

Explanations

phrases related to surprising or impactful events

oai_token-act-pair · gpt-3.5-turbo

instances of the word "shock" and its variations

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

uties

-0.82

amins

-0.82

umbers

-0.75

allery

-0.71

ittees

-0.67

 ende

-0.65

ccording

-0.64

ufact

-0.64

oug

-0.63

 notor

-0.63

POSITIVE LOGITS

waves

1.11

wave

1.10

 absor

1.04

tro

0.97

ingly

0.93

imaru

0.91

mong

0.84

er

0.83

shock

0.81

fully

0.75

Activations Density 0.016%

phrases related to surprising or impactful events

instances of the word "shock" and its variations

No Comments

No Known Activations