INDEX

Explanations

phrases related to support or assistance

oai_token-act-pair · gpt-3.5-turbo

words related to support or assistance

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

ãĥīãĥ©ãĤ´ãĥ³

-0.74

 logger

-0.70

 Hebdo

-0.69

 Nanto

-0.69

 Abyssal

-0.67

âĹ¼

-0.67

 Admir

-0.67

 Babel

-0.64

 Palm

-0.64

 Masquerade

-0.63

POSITIVE LOGITS

orter

1.16

lication

1.13

lied

1.08

orters

1.08

osition

1.07

ressive

1.06

ression

1.04

ressor

1.01

ressed

1.00

onent

0.98

Activations Density 0.015%

phrases related to support or assistance

words related to support or assistance

No Comments

No Known Activations