INDEX

Explanations

statements about likely outcomes or predictions

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate probability or likelihood of future events

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

inth

-0.86

gado

-0.81

aredevil

-0.77

zeb

-0.77

ithing

-0.75

ilts

-0.74

artney

-0.74

gencies

-0.73

gian

-0.73

ortmund

-0.71

POSITIVE LOGITS

 underest

0.81

 underestimate

0.75

 infer

0.74

 culprit

0.74

 doomed

0.72

 likely

0.70

 underestimated

0.69

 exagger

0.69

NULL

0.69

 going

0.67

Activations Density 0.036%

statements about likely outcomes or predictions

phrases that indicate probability or likelihood of future events

No Comments

No Known Activations