INDEX

Explanations

negative descriptors such as "crappy" and "shitty."

oai_token-act-pair · gpt-3.5-turbo

negative adjectives describing quality or experience

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

hip

-0.98

Topics

-0.96

mental

-0.84

ees

-0.83

encers

-0.82

irs

-0.79

Goal

-0.78

icity

-0.77

cycl

-0.76

jong

-0.76

POSITIVE LOGITS

 blah

0.97

 crappy

0.96

 crap

0.93

 shitty

0.92

 damned

0.90

 darn

0.88

 sucked

0.87

 damn

0.84

bye

0.84

 bullshit

0.82

Activations Density 0.015%

negative descriptors such as "crappy" and "shitty."

negative adjectives describing quality or experience

No Comments

No Known Activations